Invisibility Is Not Invincibility: The Impact of Intimate Partner Violence on Gay, Bisexual, and Straight Men's Mental Health.

Intimate partner violence (IPV) is a critical public health problem. However, there is limited research conducted on and about men who are survivors. This project extends previous research by examining the post-traumatic impact of diverse forms of IPV (sexual, physical, emotional, control, and stalking) on the internalized and externalized mental health of gay, bisexual, and straight men. Using data from the National Intimate Partner and Sexual Violence Survey (2011; N = 18,957), we find that all men are equally likely to report emotional victimization and controlling tactics (with between 50% and 70% doing so), while bisexual men are significantly more likely to report physical and sexual violence and gay men are significantly more likely to report intimate stalking. Due to these experiences, gay men are significantly more likely to report missing school or work, but bisexual men are significantly more likely to rate their current overall mental health as poor. Around 10% of all men, regardless of sexual orientation, report post-traumatic stress disorder symptomology and 30% of all men report difficulty sleeping. This research suggests that sexual orientation is a critical area of focus in the study of violence and mental health for men and that we can no longer ignore the voices and needs of men survivors: Invisibility is not invincibility.

Intimate partner violence (IPV) is a critical public health problem. Between 1989 and 1996, multiple agencies, both domestic (American Medical Association) and international (World Health Organization), began a campaign to end violence against women, because it was a “public health problem” and health care’s “silent epidemic” (Schornstein, 1997; World Health Organization, 1996). While there is a growing body of literature on the impact of IPV on women, there has been limited research conducted on men survivors of violence (Cascardi, O’Leary, & Schlee, 1999; Coker et al., 2002; Ellsberg, Jansen, Heise, Watts, & Garcia-Moreno, 2008; Gandhi et al., 2010; Golding, 1999; Heise, 1998; Huang, Yang, & Omaye, 2011; Hyman, Schillinger, & Lo, 1995; Peckover, 2003; Ramsay, Richardson, Carter, Davidson, & Feder, 2002; Richardson et al., 2002; Rodriguez, Craig, Mooney, & Bauer, 1998; Stark & Flitcraft, 1996; Warshaw, Ganley, & Salber, 1998; Wingood, DiClemente, & Raj, 2000). This lack of focus is unfortunate, given that the 2010–2012 summary report of the National Intimate Partner and Sexual Violence survey (NISVS) shows that in the United States 30.9% or nearly 1 in 3 men have experienced contact sexual violence, physical violence, and/or stalking by an intimate partner (Smith et al., 2017).

The impact of violence is particularly compounded for gay and bisexual men, who face IPV tactics that are specific to their sexuality—such as outing, questioning a partner’s sexual orientation (biphobia, binegativity), or coercive use of HIV status (Farrel & Cerise, 2006; Roch, Mortin, & Ritchie, 2010; Tesch & Bekerian, 2015; Turrell, 2000)—as well as barriers to post-traumatic care and service provision due to stigma and bias (Brown & Groscup, 2009; Ciarlante & Fountain, 2010; National Coalition of Anti-Violence Programs [NCVAP], 2016).

Preliminary research suggests that “lack of proper training or ability of law enforcement agencies, judges, social service workers, mental health professionals and even the media” to recognize men as victims may very well lead victims to feel like they are responsible or to blame for their own victimization (Hanna, 2015, p. 5; National Organization of Human Services database query). This is especially true in the case of sexual violence, as revealed through in-depth interviews with survivors (Donne et al., 2018). Qualitative studies find men are significantly less likely to seek professional or institutionalized help, even when controlling for their lower victimization rates (Donne et al., 2018; Douglas & Hines, 2011; Edwards, Sylaska, & Neal, 2015). According to Douglas and Hines (2011), men who sought help from hotlines, agencies, and police were often ridiculed, referred to batterers’ programs (e.g., believed to be the abusers), or simply not believed.

This disbelief and dismissal of harm is intensified for gay and bisexual men, who must also deal with structural and interpersonal homophobia. Incidents of IPV among lesbian, gay, and bisexual (LGB) partners have long been viewed as being less severe and less likely to occur than those in heterosexual relationships (Brown & Groscup, 2009). Because of this, services for lesbian, gay, bisexual, transgender, queer and otherwise not-heterosexual, not-cisgender (LGBTQ+) survivors are severely lacking. A study by the NCAVP and the National Center for Victims of Crime found that 94% of domestic violence agencies, sexual assault centers, prosecutors’ offices, law enforcement agencies, and child victim services providers said they were not serving LGBTQ survivors of IPV and sexual violence (Ciarlante & Fountain, 2010). Making things worse, NCAVP reported that in 2015, 44% of LGBTQ+ survivors of IPV were denied shelter services, and of those who interacted with the police (most did not), 12% reported that the police were hostile, 13% said that the police were indifferent in their actions, and 31% said they experienced misarrest, meaning the survivor was arrested rather than the abusive partner (this is up from 17% in 2014; NCAVP, 2016).

Qualitative studies and small surveys confirm that these barriers often produce a “double closet” for LGBTQ+ people, that is, feeling the need to keep secret not only their sexual orientation or intimate relationships but also the abuse and trauma they are experiencing (McClennen, 2005; St Pierre & Senn, 2010). This double closet has numerous consequences for men’s mental health. For instance, men who have experienced trauma in their lives report not only increased self-injury as a means of attempting to cope with pain (including drinking and use of drugs) but also internalizing behaviors (such as attempting to fix their problems alone) and feelings of isolation (Oliffe et al., 2017). One recent review of the literature found that gay men who have experienced trauma face additional isolating factors related to familial and peer rejection, lack of social community, and internalized stigma, which all intensify feelings of depression and suicidality (Lee, Oliffe, Kelly, & Ferlatte, 2017).

More recently, small surveys have examined the most common psychological outcomes triggered by IPV in men, such as alcohol abuse, substance abuse, post-traumatic stress disorder (PTSD; Caldwell, Swan, & Woodbrown, 2012; Douglas & Hines, 2011), suicidality (Golding, 1999), and depression (Caldwell et al., 2012; Chang et al., 2010; Hines & Malley-Morrison, 2001; Richards, Noret, & Rivers, 2003). Coker et al. (2002), examining the National Violence Against Women Survey, highlighted the need to more critically examine the psychological impact of psychological, emotional, and control violences specifically because these are the most commonly reported forms of abuse men experience. Nowinski and Bowen (2012) confirm these forms of abuse are commonly experienced by gay men, in a meta-review of the literature. One early qualitative study found that men who had experienced physical assault by an intimate partner were significantly more likely to meet criteria for PTSD than men who had been physically assaulted by someone other than an intimate partner (e.g., by a stranger, in their workplace, by a parent; Dansky, Byrne, & Brandy, 1999). A reanalysis of the Canadian General Social Survey data by Laroche (2005; N = 25,876) finds that 83% of men who “feared for their life” did so because they were unilaterally terrorized by their intimate partner. Of these terrorized men, 80% reported having their everyday activities disrupted.

The findings on the health-related consequences of IPV for gay and bisexual men are, however, limited because an overwhelming majority of research on LGB IPV focuses on lesbian and bisexual women or examines men and women together due to small sample size issues—making it difficult to parse out nuanced differences in experiences (Coston, 2017; Edwards, Sylaska, & Neal, 2015). For all LGB people studied together, qualitative studies confirm that experiencing IPV elevates the risk for mental health issues such as depression and anxiety (Campbell, 2002; Coker et al., 2002; Pico-Alfonso et al., 2006; Rennison & Welchans, 2000), while quantitative studies find that an overwhelming majority of all victims of IPV report low satisfaction with life and poor perceived health, and almost half report more than 7 poor mental health days a month (Blosnich & Bossarte, 2009). A more recent quantitative study found that LGB survivors of violence were significantly more likely than non-LGB survivors to report both depression and anxiety, specifically (Miller & Irvin, 2017).

Results of these studies also vary depending on sample used (survey, hotline, emergency room) and whether the data is population based or not. For instance, smaller, qualitative studies often find a higher proportion of men reporting post-traumatic mental health issues (Shorey et al., 2011) than larger population-based surveys do (Ehrensaft, Moffitt, & Caspi, 2006; Smith et al., 2017; Tjaden & Thoennes, 2000).

Given the above, this project extends previous research in five key ways: (a) It is the first project to quantitatively test for post-IPV mental health disparities between self-identified heterosexual, gay, and bisexual men, separately; (b) it is the first project to do so using population-based data; (c) it is the first project to also use specific key mental health indicators including both a general, global self-perception measure and specific outcomes such as PTSD, difficulty sleeping, or missing school/work; (d) it is the first project to look not at lifetime victimization—which often includes peers, acquaintances, and family members—but at adult IPV, defined here as it is in studies of gender-based violence against women; and (e) it is the first project to examine multiple, diverse forms of IPV: physical, sexual, psychological, emotional/controlling tactics, and stalking. In doing so, it follows Randle and Graham’s (2011) suggestions to conduct research on men’s postvictimization psychological health outcomes that uses validated, diverse IPV measured; is cross-sectional, with a large sample of men; includes both internalizing (e.g., self-perceived mental health status, PTSD) and externalizing (e.g., difficulty sleeping and missing school/work) psychological measures; and includes sexual orientation disparities, a critically needed area of research that Randle and Graham indicate is in need of further development.

Methods

Data come from the Centers for Disease Control and Prevention’s NSIVS, an ongoing nationally representative random-digit-dial telephone survey of the noninstitutionalized English- and Spanish-speaking U.S. population aged ≥18 years. The survey protocol received approval from the Office of Management and Budget (OMB# 0920-0822) as well as the Institutional Review Board of the Research Triangle Institute, International. The use of this dataset for secondary data analyses was approved in 2017 by Virginia Commonwealth University’s Institutional Review Board.

Sample

In this dataset there were 3,623 heterosexual-/straight-identified men, 142 gay-identified men, and 88 bisexually identified men. While these sample sizes might seem uneven, the Williams Institute estimates that only 3.5% of the population identifies as lesbian, gay, or bisexual, with gay men comprising substantially more than half of gay and bisexual men on a majority of large surveys (Gates, 2011). While a more recent Gallup poll finds the size of the LGBT population to be 3.9% for men, this includes transgender people—who make up around 0.5% of the population (Newport, 2018). In this way, we might reasonably conclude that around 3.5% of men identify as gay and bisexual. In this survey, 6.35% of the sample identifies as such.

This analysis examines the negative mental health outcomes of physical, sexual, emotional and control, and stalking IPV. As opposed to including all forms of IPV—such as violence by an acquaintance, coworker, and/or family member—this analysis only considers intimate partners to be current or former (including ex) husbands, wives, live-in partners, fiancés, boyfriends, girlfriends, dating partners, “someone you were seeing,” and/or “someone you were having sex with,” thus making this analysis directly comparable to a majority of the previous IPV research on women abused by men. The violence variables were constructed using the questions and survey scale items developed and rigorously tested by NISVS researchers in the following ways: Scales were taken from the National Violence Against Women Survey (1995–1996), a federal workshop that focused on building data systems for monitoring and responding to violence, pilot studies conducted in 2007, an expert panel that reviewed those pilot findings, and cognitive testing of the questionnaire on diverse participants (United States Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2016).

Physical violence

Physical violence was measured with a series of questions, including whether a romantic or sexual partner had ever “slapped you; pushed or shoved you; hit you with a fist or something hard; kicked you; hurt you by pulling your hair; slammed you against something; tried to hurt you by choking or suffocating you; beaten you; burned you on purpose; and/or used a knife or gun on you.”

Sexual violence

Sexual violence was measured by asking if romantic or sexual partners had ever “exposed their sexual body parts to you, flashed you, or masturbated in front of you; made you show your sexual body parts to them when you didn’t want it to happen; made you look at or participate in sexual photos or movies; fondled, groped, grabbed, or touched you in a way that made you feel unsafe”; or when drunk, high, passed out, or unable to consent, ever “had vaginal sex with you; made you perform anal, oral, or vaginal sex” (either forcing themselves on you or forcing you to penetrate/have sex with them); or “forced you to engage in sexual activity.” This was a question assessed within the subset of physical violence questions.

Psychological violence

Psychological violence was measured by asking whether a romantic or sexual partner had ever “acted very angry towards you in a way that seemed dangerous; told you that you were a loser, a failure, or not good enough; called you names like ugly, fat, crazy, or stupid; insulted, humiliated, or made fun of you in front of others; told you that no one else would want you.”

Control violence

Control violence was measured by asking whether romantic or sexual partners had ever “tried to keep you from seeing or talking to your family or friends; made decisions for you that should have been yours to make, such as the clothes you wear, things you eat, or the friends you have; kept track of you by demanding to know where you were and what you were doing; threatened to hurt someone you love or did hurt someone you love; said things like ‘If I can’t have you, then no one can’”; or, if applicable, “tried to get you pregnant when you did not want to become pregnant; [if male] tried to get pregnant when you did not want them to get pregnant; tried to stop you from using birth control; refused to use a condom when you wanted them to use one.”

Intimate stalking

Stalking included the following behaviors from romantic and/or sexual partners: “watched or followed you from a distance, or spied on you with a listening device, camera, or GPS [global positioning system]; approached you or showed up in places, such as your home, workplace, or school when you didn’t want them to be there; left strange or potentially threatening items for you to find; sent you unwanted emails, instant messages, or sent messages through websites like MySpace or Facebook; and/or left you cards, letters, flowers, or presents when they knew you didn’t want them to.”

Dependent Variables

Follow-up questions were asked of all those reporting physical, sexual, emotional, psychological/control, and/or intimate stalking violence to assess the impact that violence has had on their lives and well-being. The dependent variables were constructed by examining the literature on IPV and anxiety, depression, and PTSD (as cited in the following text).

Difficulty sleeping

Difficulty sleeping was measured if a person responded affirmatively to the question “Have you ever been told by a doctor, nurse, or other healthcare professional that you had . . . difficulty sleeping?” Research indicates that difficulty sleeping is one of the key chronic conditions reported by heterosexual/straight women who have been victimized. Indeed, Campbell and Lewandowski (1997) found that “battered women generally would not complain of PTSD per se to a health care provider, but rather of sleep disorders or stress.” Thus, there is a substantial probability of misdiagnosis or lack of diagnosis of PTSD by non–mental health providers (Hoffman & Paradise, 2008).

Missing work or school

Missing school or work was measured if a person responded affirmatively to the question “Did you ever have to miss days of work or school when this/any of these things happened with [the abuser]?” This measure was included, as research indicates that exposure to IPV can greatly disrupt an individual’s ability to maintain employment or regular attendance in school, typically due to the inability to cope with stressors and some of the associated co-occurring aftereffects (such as alcohol use, depression, and anxiety; Browne, Salomon, & Bassuk, 1999).

Post-traumatic stress disorder

To measure the prevalence of postvictimization psychological distress, a binary variable was created so that if you did experience PTSD after victimization you scored 1 and if you did not, you scored 0. Questions were written into the survey itself based on their alignment with the PTSD Symptom Scale, a 20-item self-report scale based on the Diagnostic and Statistical Manual of Mental Disorders (5th ed., DSM-V; American Psychiatric Association, 2013) criteria (Foa et al., 2016). In order to count as PTSD, an individual needed to affirmatively respond “yes” to at least one reexperiencing/intrusion symptom (in this case “yes” to having nightmares about the victimization), one avoidance symptom (“yes” to going out of the way to avoid thinking about the victimization), two changes in cognition and/or mood symptoms (“yes” to both feeling numb or detached from others and missing school/work because of how it impacted them), and two increased arousal and reactivity symptoms (“yes” to being concerned for their safety, fearful, and/or constantly on guard).

Current self-rated overall mental health

The subjective measure of current overall mental health was measured using the question “Would you say that your overall mental health is . . . ” (a) excellent, (b) very good, (c) good, (d) fair, or (e) poor. Excellent, very good, and good were collapsed into a single category of good mental health (0), while fair and poor were collapsed into a single category of poor mental health (1). Subjective or self-assessed measures of mental health are often used to study the impact of IPV victimization; indeed Dillon, Hussain, Loxton, and Rahman (2013) found that of the 62 peer-reviewed studies on IPV and mental health outcomes between 2008 and 2012, 19 (31%) used some form of subjective of self-assessed measure of psychological well-being or distress.

Combined negative mental health measure

Answers to the variables already mentioned were also merged to create a scale variable ranging from 0—no experiences of postvictimization negative mental health outcomes or status—to 8—experiencing the full range of postvictimization negative mental health consequences. This includes now separate measures for difficulty sleeping, missing school or work, and PTSD symptomology. Thus, this measure includes the following: difficulty sleeping (0 = no, 1= yes), PTSD symptomology (0 = no, 1 = some but not all, 2 = diagnosable symptomatology), missing school or work (0 = no, 1 = yes), self-rated overall mental health (0 = excellent, 1 = very good, 2 = good, 3 = fair, and 4 = poor), and the postvictimization experience.

Statistical Analyses

This study uses Stata/MP 14 quantitative data analysis software for large datasets (applying standardized sample weights and adjusting standard errors for both clustering and stratification using SVY for survey data) to complete a series of χ2 tests and an analysis of variance (ANOVA) test, to determine if there are significant differences between one or more of the groups in negative mental health outcomes.

Chi-Square Analyses

Chi-square tests determine if there are significant differences between one or more of the groups in their postvictimization mental health outcomes. To assess which group is contributing most to any significant results, adjusted standardized Pearson residuals were computed. According to Agresti, “A[an adjusted] standardized residual having absolute value that exceeds about 2 when there are few cells or about 3 when there are many cells indicates lack of fit of Ho in that cell” (1997, p. 38). Note that F statistics are reported instead of x2 statistics because of the survey-weight set data. This method has been used by others studying LGBT IPV and its health-related consequences (Coston, 2018, 2019).

ANOVA analysis

Analysis of variance tests determine if there are significant mean differences between three or more groups in the postvictimization cumulative negative mental health measure. Postestimation groupwise Wald tests are computed to compare gay men to straight men, gay men to bisexual men, and straight men to bisexual men, separately, to assess which mean scores are contributing to the overall significant ANOVA. However, postestimation effect sizes (such as Cohen’s d) are not possible for more than two groups with survey set data, so it’s not possible to measure whether or not these significant mean differences are meaningfully small, medium, or large.

Results

Table 1 shows the distribution of men in each of these self-identified sexual orientation categories who experienced IPV of any kind (e.g., physical, sexual, emotional, psychological/control, and/or stalking) by a man or a woman. We see here that bisexual and gay men experience statistically significantly higher rates of IPV than heterosexual/straight men; specifically, while nearly 79% (N = 82) of gay men and more than 83% (N = 54) of bisexual men report IPV, only 64% (N = 1,952) of straight men do. Though it’s important to point out here that these rates are much higher than many previous studies have found, which we conclude is due to our expansion of violence types and our defining an intimate partner as a current or former dating, romantic, or cohabiting partner (as described in more detail in the following text), it is also important to note that while a majority of straight men experience violence at the hands of an intimate woman partner, nearly one third (27.2%) report violence by an intimate male partner. The results are mirrored for bisexual men, with one third reporting violence by women (33.3%) and two thirds by men (67.7%). Finally, only 3.66% of gay men report violence from an intimate woman partner.

Table 1.

Percent of Men Experiencing Any Form of Intimate Partner Violence.

	%	n	Pearson Residual
Heterosexual/straight men (N = 3,465)	63.84	1952	−0.724
Proportion woman partner	72.80
Proportion man partner	27.20
Bisexual men (N = 84)	83.33	54	2.106
Proportion woman partner	3.66
Proportion man partner	96.34
Gay men (N = 133)	78.95	82	2.022
Proportion woman partner	33.33
Proportion man partner	66.67
	F(1.89, 190.44) = 11.0549, p = .001

Note. The percentage listed here represents the number of men (N) who experience violence out of the total number of men within that particular sexual orientation category. For instance, 63.84% of all heterosexually identified men experienced intimate partner violence, or 2,212 out of 3,465 men.

Straight, bisexual, and gay men are equivalently likely to experience emotional and control violence, with around 60%–70% reporting emotional victimization and 70%–80% reporting controlling tactics by abusers (Table 2). However, gay and bisexual men are significantly more likely to report sexual violence victimization (F(1.97, 198.98) = 9.9612, p > .001), while bisexual men are significantly more likely to report physical violence (F(1.97, 199.01) = 3.214, p > .05) and gay men are significantly more likely to report being stalked by an intimate partner (F(1.97, 198.54) = 10.2657, p > .001). Specifically, while two thirds of straight men report sexual violence victimization, nearly 50% of gay men and 60% of bisexual men do so; and while 50%–60% of straight and bisexual men report being stalked, over 66% of gay men do so.

Table 2.

Percent of Men Experiencing Intimate Partner Violence by Type.

	Sexual			Physical			Emotional			Control			Stalking
	%	n	Pearson Residual	%	n	Pearson Residual	%	n	Pearson Residual	%	n	Pearson Residual	%	n	Pearson Residual
Heterosexual/straight men (N = 1,952)	34.48	673	−0.877	59.58	1163	0.045	61.63	1203	–	81.45	1590	–	46.52	908	−0.728
Bisexual men (N = 54)	59.26	32	2.904	70.37	38	2.365	57.41	31	–	79.63	43	–	60.38	32	1.342
Gay men (N = 82)	48.19	40	1.911	50.60	42	−1.876	71.08	59	–	74.70	62	–	66.27	55	2.456
F	9.9612			3.214			1.434			1.2885			10.2657
p	0.0001			0.0423			0.2408			0.2777			0.0001

Table 3 shows the mixed results of the χ2 tests. For instance, there is no reported statistically significant difference between heterosexual, bisexual, and gay men in postviolence difficulty sleeping (F(1.90, 191.48) = 1.3453, p > .05) or PTSD symptomology (F(1.94, 190.94) = 0.3793, p > .05): Around one third of all men report difficulty sleeping and around 5%–10% report symptoms that diagnostically fit PTSD (such as mood or cognitive disturbances, reactivity, or intrusive thoughts). However, there is a significant difference between men and their missing school or work due to victimization and their current self-rated overall mental health. Specifically, we can see that over 20% of gay men report post-traumatic missed school or work (F(1.99, 201.11) = 3.4792, p < .05), compared to 10%–14% of straight and bisexual men. While it is bisexual men who are more likely to report poor overall current mental health (F(1.98, 199.61) = 5.3225, p < .01), over 30% of bisexual men rate their current mental health as fair or poor, in comparison to 14% of straight men and 22% of gay men.

Table 3.

Chi-Square Tests for Significant Differences in Percent of Individuals Reporting Various Negative Mental Health Outcomes, by Sexual Orientation.

	Difficulty Sleeping			Missed School or Work			Post-Traumatic Stress Disorder Symptomology			Current Self-Rated Mental Health is Poor or Fair
	%	n	Pearson Residual	%	n	Pearson Residual	%	n	Pearson Residual	%	n	Pearson Residual
Heterosexual/s\traight men (N = 1,952)	28.91	564	–	13.76	255	−0.321	8.4	40	–	14.32	210	−0.760
Bisexual men (N = 54)	27.78	15	–	9.43	5	−0.895	12.2	5	–	32.50	13	2.833
Gay men (N = 82)	37.80	31	–	23.46	19	2.261	5.56	1	–	21.88	14	1.396
	F(1.90, 191.48) = 1.3453, p > .05			F(1.99, 201.11) = 3.4792, p < .05			F(1.94, 190.14) = 0.3793, p > .05			F(1.98, 199.61) = 5.3225, p < .01

Note. (a) Proportion (percentage) of those experiencing mental health impact is reported in first column; total number within sexual orientation group reported in second column; adjusted standardized Pearson χ2 residual, measuring individual cell contribution to overall ×2, in third column. (b) Post-traumatic stress disorder (PTSD) N is smaller than the other two measures as fewer men answered these follow-up questions.

In order to get a general sense of the combined impact of these outcomes, an additional comparison test was conducted, looking at the cumulative number of affirmative responses to difficulty sleeping, missing school or work, PTSD symptomology, and poor current self-rated mental health status. Table 4 shows that bisexual and gay men score significantly higher, on average, than straight men do (F(2, 101) = 3.63, p < .05). Specifically, straight men average a score of 1.81 out of a possible high score of 8, while gay men average 2.37 and bisexual men average 2.26 out of 8. The 95% confidence intervals for gay [1.929, 2.818] and bisexual men [1.782, 2.736] overlap quite substantially, but neither overlaps with those for straight men [1.982, 1.899]. This indicates that there is likely not a meaningful difference between gay and bisexual men’s mean scores (post hoc estimate tests confirm this).

Table 4.

Mean Score Postvictimization Negative Mental Health Consequences, by Sexual Orientation.

	Mean Score	SD	95% CI
Heterosexual/straight	1.831	0.032	[1.762, 1.899]
Bisexual	2.259	0.211	[1.982, 2.736]
Gay	2.373	0.184	[1.929, 2.818]
Wald F	3.63
p	.03

Note. Negative mental health was a scaled variable taking into account reports of current difficulty sleeping, missing school or work, self-reported overall current mental health status, and some post-traumatic stress disorder (PTSD) symptomology. Scores ranged from 0 (no reported negative mental health consequences) to 8 (difficulty sleeping, missing school/work, poor self-rated health, PTSD).

When comparing groups using postestimation tests, both gay men and bisexual men have significantly higher average scores than straight men. However, gay and bisexual men do not significantly differ from each other in their mean scores.

Discussion

The results in the preceding text have not been previously reported and confirm what some other, smaller and nonprobability studies have found: that gay and bisexual men are, generally, significantly more likely to report violence and its negative mental health outcomes. However, there were three key findings that add nuance and depth to our general understandings. The first is that gay men are nearly two times more likely than both bisexual and straight men to report missing school or work due to IPV victimization. Here, we posit that this is linked to gay men also being significantly more likely to report that they’ve been stalked by their intimate partner. While intimate stalking is often excluded from large studies on IPV, it is nonetheless an important area due more consideration. Indeed, no fewer than half of the men in this study reported being intimately stalked. And, perhaps unlike other forms of violence, stalking very directly impacts one’s ability to leave the house and feel safe; although we note that there are psychological consequences to all forms of violence that impede people’s ability to leave the house, here there is the potential for both generalized fear or anxiety around leaving the house and targeted anxiety around seeing or having to confront your stalker. Indeed, research of women’s experiences of being stalked commonly discusses changes to routines, activities, and behaviors, including avoiding places where the stalker could find you or know you might be (Amar, 2006; Sinwelski & Vinton, 2001). Research on women’s experiences of being stalked also shows that stalking victimization impacts workplace performance, absenteeism, tardiness, and on-the-job interference tactics including asking victims to leave their job immediately or verbally harassing coworkers or a supervisor (Swanberg & Logan, 2005). As a majority of states do not have employment nondiscrimination policies, it’s incredibly important to further study the impact that stalking has on gay men’s ability to continue school and work and perhaps more deeply on gay men’s experiences of being stalked at work and the impact that has on their employment status.

The second key finding is that bisexual men are significantly more likely than straight and gay men to rate their current overall mental health as fair or poor. This supports previous research on mental health among bisexual individuals, which indicates that bisexual people suffer the double impact of homophobia and binegativity, leading to much higher levels of minority stress than gay and lesbian people (Lewis, Derlega, Brown, Rose, & Henson, 2009). Specifically, previous research has confirmed that sexual orientation–related stigma and prejudice are especially amplified for bisexual individuals, who face the burden of exclusion, rejection, and marginalization from not only heterosexuals and heteronormative society but also from lesbians/gay individuals, and are often stereotyped as having illegitimate or unstable sexualities or being sexually promiscuous or irresponsible (Balsam & Mohr, 2007; Bradford, 2004; Erickson-Schroth & Mitchell, 2009; Herek, 1997, 2002; Koh & Ross, 2006; Mohr & Rochlen, 1999; Mulick & Wright Jr., 2002). When coupled with the fact that bisexual individuals are also significantly more likely to experience IPV victimization in their lifetimes (Coston, 2017, 2018) and that this article finds bisexual men are significantly more likely to experience physical and sexual IPV, specifically, it’s not surprising that they would be more likely to rate their current mental overall mental health poorly.

Finally, when taken together, gay and bisexual men report a higher total number of postvictimization negative mental health outcomes, on average, than straight men do. As we discussed earlier, this could be due to the double impact of victimization and minority stress. When taking into consideration the downplaying, disbelief, denial of services, the general social perception of men as brave, stoic, strong, and, in some cases, invincible, and the stigma associated with nonheterosexual sexualities, gay and bisexual men experience an intense buildup of post-traumatic guilt, shame, fear, and anxiety. Indeed, if men who are survivors are constantly told that they cannot be victimized based on society’s expectations of them, they are more likely to suffer high levels of internal conflict.

Unfortunately, standard procedures for screening men for IPV within health settings have yet to be established and qualitative studies show that IPV is most effectively screened when there is perceived high risk for the patient (Snider, Webster, O’Sullivan, & Campbell, 2009; Witting et al., 2006). In many health settings, the health-related indicators that often give away the presence of IPV in women (e.g., broken bones and other bodily injuries, the appearance of low self-esteem, a history of alcohol or drug abuse, and a history of anxiety, depression, or suicide attempt) are often explained away as behavioral or general mental health problems in men (e.g., as a symptom of masculinity itself; Chuick et al., 2009; Cochran & Rabinowitz, 1999). The “boys will be boys” mentality potentially makes it “easier” for men to hide or explain away the post-traumatic symptoms and scars of IPV, while the erasure of nonheterosexuality makes health-care providers themselves less likely to screen for IPV in LGB people—and particularly gay and bisexual men (Calton, Cattaneo, & Gebhard, 2016; Douglas & Hines, 2011; Messinger, 2011; Shakil, Donald, Sinacore, & Krepcho, 2005; Stephenson & Finneron, 2013).

Limitations

It is important to note that there are still several limitations to this study that could have an effect on the results and their interpretation. First, the sample of men who identified themselves as gay or bisexual was equivalent to 6.35% of the total sample size. Although this is a larger proportion of gay and bisexual men than is found in the general population, it is still a small sample and it impacted our inability to perform more complex statistical analyses, such as regression analyses in which we examined IPV and its outcomes along intersecting axes (e.g., age, income, race ethnicity, disability). Another limitation to the study is the possibility of misclassification of gay and bisexual men within the chosen population study. A recent study conducted by Ferlatte, Hottes, Trussler, and Marchand (2017) explored misclassification bias and it was concluded that nearly one third of men who participated in the study would not be willing to reveal their sexual orientation on a government study. While this was not marketed as a government survey, we must still be careful as the conclusions here may be an underrepresentation of particular lived experiences.

Conclusion

This is the first study using this data source to analyze disparities in mental health outcomes between straight, gay, and bisexual men. However, it is important that we note this study in no way undermines or detracts from the critically necessary work examining women’s experiences of IPV. We believe that while structural resources for serving survivors of violence are limited and under constant threat of dissolution (e.g., debates over renewal of the Violence Against Women Act), these barriers should not prevent us from also discussing the real and pressing needs of gay and bisexual men who are also survivors of violence.

Indeed, the results of this study suggest that sexual orientation is a critical area of focus in the study of violence and mental health for men. Bisexual men are significantly more likely than gay and straight men to experience physical and sexual violence and gay men are significantly more likely than bisexual and straight men to experience intimate stalking. What’s more, gay men are more likely to miss school or work because of this victimization and bisexual men are more likely to report their overall mental health is poor—likely due to both minority stress and their experiences with physical violence and sexual assault. However, it’s also important to note that in this study, no fewer than one third of men report experiencing violence and an overwhelming majority report emotional violence and controlling tactics. As such, our efforts to end violence and support victims should not only include serious and critical conversations about men as survivors but also align with best practices for sexuality-based competencies in health-care provision.

Men are often overlooked in IPV research, clinical screenings, workshops and trainings, and other settings that have historically centered on understanding of violence perpetration that inherently excludes men as victims/survivors. As this study shows, this is not because they do not experience violence, and neither is it because when they do there is no consequence to them or their health. It appears as though presumed invincibility is a kind of invisibility for men, particularly gay and bisexual men.

Future research should attempt to better explain the connection between violence, mental health, and sexuality. For instance, survey items should be constructed to specifically measure and confirm the role of gender-based power, relational inequalities, and binegativity, issues we can infer here, but not directly test. What’s more, it will be critical to examine the health-care-seeking behaviors of men survivors, to discern if their needs are being met and how their experiences with health-care professionals take shape. Clinically, it is important for health-care providers to be aware of the significant impact of IPV on the men they regularly provide care to, especially as it pertains to men’s overall sense of health and well-being, ability to continue working or going to school, and issues of fear, anxiety, or sleeplessness. Men already avoid seeking health care and/or admitting they are having health problems, so we must be diligent in providing post-traumatic care to them (Smith, Braunack-Mayer, & Wittert, 2006).

In sum, we need more extensive research on men’s lived experiences as survivors so that our trauma-informed care approaches ensure well-rounded, quality, and culturally competent care for all people.