Minority Stress and Intimate Partner Violence among Sexual and Gender Minorities assigned Female at Birth

Abstract

Objective:

Sexual and gender minorities assigned female at birth (SGM-AFAB) experience high rates of intimate partner violence (IPV). Using multiwave longitudinal data, the present study tested the following associations of minority stress and IPV among SGM-AFAB: concurrent within-person (i.e., whether changes in minority stress co-occur with changes in IPV), prospective within-person (i.e., whether changes in minority stressors precede changes in IPV), and between-persons (i.e., whether individuals who experience more minority stress, on average, experience more IPV).

Method:

Data were from Waves 1–7 (spanning 3.5 years) of a longitudinal cohort study of 488 young SGM-AFAB. At each wave, participants reported on SGM victimization, sexual orientation microaggressions, internalized heterosexism, and five types of IPV for up to three partners in the past six months.

Results:

Controlling for other minority stress experiences, microaggressions showed concurrent within-person associations with two types of IPV victimization (psychological and coercive control) and three types of IPV perpetration (psychological, physical, and sexual), and between-persons associations with psychological IPV victimization and coercive control victimization and perpetration. Microaggressions also had a significant prospective within-persons association with SGM-specific IPV perpetration. SGM victimization showed no unique within-person associations with IPV but, between-persons, was associated with all types of IPV victimization (except coercive control), and psychological and sexual IPV perpetration. Internalized heterosexism was not associated with any IPV outcome.

Conclusions:

Minority stress, particularly distal stressors, appear to contribute to risk for IPV among SGM-AFAB. Clinicians working with SGM-AFAB individuals in relationships may benefit from screening for experiences of minority stress.

Intimate partner violence (IPV), defined as physical violence, sexual violence, stalking and psychological aggression (including coercive tactics) by a current or former intimate partner, is a significant public health problem that affects about 1 in 4 women in the United States (Breiding et al., 2015). IPV carries with it a host of negative health consequences, including physical injury; mental health problems including depression and posttraumatic stress disorder; increased risk for smoking, binge drinking, and risky sexual behaviors; and death (Niolon et al., 2017). Exact prevalence rates are difficult to estimate due to methodological differences across studies (e.g., how IPV is defined and measured, time period to which the violence corresponds, the relationship being studied, the definition of sexual orientation). However, a recent literature review suggests that rates of IPV among sexual minority individuals are similar to or greater than that of heterosexuals (Edwards et al., 2015). Furthermore, a growing body of research suggests that sexual and gender minorities assigned female at birth (SGM-AFAB) may be at greatest risk for IPV. For example, one longitudinal study found that the odds of physical victimization were 75% higher for SGM-AFAB than SGM assigned male at birth (Whitton, Newcomb, et al., 2019).

To understand the etiology of increased rates of IPV among SGM-AFAB, it is important to investigate unique factors associated with IPV in this population. One of the most widely studied risk factors is sexual minority stress (Brooks, 1981). Meyer (2003) defined minority stress as the unique stress derived from being a member of a socially stigmatized group. These stressful experiences exist on a continuum from distal to proximal. Distal minority stressors are defined as external, objective stressful events and conditions (e.g., victimization, microaggressions). On the other end of the continuum are proximal minority stress processes, which include the internalization of negative societal attitudes (i.e., internalized heterosexism; Meyer, 2003). A small but growing body of research supports links between minority stress experiences and various forms of IPV among SGM-AFAB. In terms of distal minority stressors, two studies of female same-sex couples found that experiences of discrimination were associated with IPV perpetration and victimization (Balsam & Szymanski, 2005; Do et al., 2022). In regard to more proximal minority stress experiences, internalized heterosexism has been associated with psychological, physical, and sexual IPV victimization and perpetration among women in same-sex relationships (Balsam & Szymanski, 2005; Lewis et al., 2014; Milletich et al., 2014).

The goal of the present study was to use longitudinal data to investigate within- and between-persons associations of minority stress with IPV victimization and perpetration among SGM-AFAB. This study will address several limitations in existing research. First, previous measurement of IPV may not have fully captured the IPV experiences of SGM. Although sexual minorities experience the same types of IPV that heterosexuals do (e.g., physical, sexual), they also experience unique forms of IPV related to their marginalized status (e.g., threatening to “out” one’s partner at work; Burke et al., 2002). Yet, according to meta-analyses and systematic literature reviews of IPV in same-sex relationships, this type of SGM-specific IPV has rarely been assessed (Badenes-Ribera et al., 2016; Edwards et al., 2015; Lewis et al., 2012; Longobardi & Badenes-Ribera, 2017). In addition, psychological IPV, despite being more prevalent than physical IPV in same-sex relationships, has been measured far less frequently (Longobardi & Badenes-Ribera, 2017). It is crucial to assess psychological IPV, given its higher prevalence and potential to be more detrimental to mental health than physical aggression (Hellemans et al., 2015; Pepper & Sand, 2015). Further, coercive control, defined as nonphysical coercion, demands, or constraints in the context of a credible threat of consequences from noncompliance, is under-researched among SGM-AFAB, despite evidence that it is an important form of IPV victimization among women (Catallozzi et al., 2011; Dutton & Goodman, 2005; Miller, 2006).

There are also gaps in the IPV literature in assessment of minority stress. Previous studies examining distal minority stress have focused on acute, overt acts of heterosexist discrimination (e.g., being disowned by one’s family, being fired from one’s job, being physically attacked). However, it has been argued that, given shifting societal attitudes toward sexual minorities, discrimination has taken on more subtle and covert forms in contemporary society (Sue et al., 2007). Indeed, sexual minorities report experiencing microaggressions, defined as daily environmental, behavioral, and verbal occurrences that convey negative messages to marginalized individuals, much more commonly than acute violent events (Kosciw et al., 2016). Heterosexist microaggressions have been associated with negative outcomes such as mental health problems (e.g., Swann, Minshew, Newcomb, & Mustanski, 2016). However, to our knowledge, only one study has examined associations of heterosexist microaggressions with IPV among racial minority SGM-AFAB and found that they were associated with increased physical, psychological, sexual, and SGM-specific IPV (Swann et al., 2021). Additional research is needed to evaluate the within-persons associations of microaggressions and IPV among SGM-AFAB.

In addition, most previous research on minority stress and IPV among SGM-AFAB has only focused on a subset of this population, namely, cisgender women in relationships with cisgender women (e.g., Balsam & Szymanski, 2005; Do et al., 2022; Lewis et al., 2014; Milletich et al., 2014). SGM-AFAB include sizeable numbers of gender minorities (Salk et al., 2020), as well as those who are in relationships with cisgender men and gender minority individuals (Bauermeister et al., 2010; Dyar et al., 2020). To be truly representative of this diverse population, samples of SGM-AFAB must be inclusive of gender minority individuals (e.g., transgender men, nonbinary individuals assigned female at birth) and those who are in same-sex and mixed-sex relationships.

Lastly, because previous research on minority stress and IPV has been largely cross-sectional, it is limited in its ability to contribute evidence that minority stress and IPV covary over time within individuals, and that increases in minority stress may cause increases in risk for IPV. In contrast, the present study examined within-persons associations between minority stressors at one wave and IPV at the same wave (concurrent associations) and at the next wave (prospective associations). Concurrent within-persons associations test whether changes in an individual’s minority stress experiences tend to co-occur with changes in their likelihood of IPV, whereas prospective within-persons associations assess whether changes in minority stressors tend to precede changes in their likelihood of IPV. The findings of this study have the potential to provide strong evidence that interventions to reduce minority stress may, in turn, reduce risk for IPV among SGM individuals in relationships.

Present Study

The present study used data from a longitudinal cohort study of young SGM-AFAB to investigate associations of minority stress with IPV victimization and perpetration. To address gaps in previous research, we assessed the occurrence of five types of IPV victimization and perpetration: physical, psychological, sexual, coercive control, and SGM-specific. In addition, we assessed distal (acute, overt acts of SGM victimization and more covert microaggressions) and proximal (internalized heterosexism) minority stress variables. We used multiple waves of data, collected at 6-month intervals (with the exception of the last wave, which was collected one year after the previous wave) over a span of 3.5 years. We examined within-persons (concurrent and prospective) and between-persons associations between minority stressors and IPV.

We hypothesized that at waves when individuals experienced more minority stress (i.e., SGM victimization, sexual orientation microaggressions, and internalized heterosexism), they would have a higher likelihood of experiencing IPV (i.e.., psychological, physical, sexual, and SGM-specific IPV, and coercive control) victimization and perpetration at the same wave (i.e., within-persons concurrent associations; Hypothesis 1), as well as at the following wave (i.e., within-persons prospective associations; Hypothesis 2). In addition, based on previous cross-sectional research showing that individuals who experience more minority stress report higher rates of IPV, we hypothesized that individuals who had higher levels of minority stress on average across waves would also have a higher likelihood of experiencing IPV victimization and perpetration (i.e., between-persons association; Hypothesis 3).

Method

Participants

The sample consisted of 488 SGM-AFAB participating in FAB 400, a longitudinal cohort study that began in November 2016. To achieve a multiple cohort, accelerated longitudinal design, FAB 400 includes SGM-AFAB from two cohorts: (1) a late adolescent cohort recruited for FAB 400 in 2016–2017 (N = 400; 16- to 20-years-old at baseline); and (2) a young adult cohort comprised of the AFAB participants from Project Q2, a previous cohort study of SGM youth recruited in 2007 (N = 88; 23- to 32-years old at the FAB 400 baseline). Eligibility criteria at original cohort enrollment were being ages 16–20 years old, assigned female at birth, and either identifying with a sexual or gender minority label or reporting same-sex attractions or sexual behavior. Participant demographics are presented in Table 1.

Table 1.

Participant Demographics (N = 488)

	M (SD)
Age	20.06 (3.66)
	N (%)
Race/Ethnicity
Black/African American	170 (34.8)
White	127 (26.0)
Hispanic or Latino/Latina/Latinx	120 (24.6)
Multiracial	42 (8.6)
Asian	25 (4.9)
Other	5 (1.0)
Sexual Orientation
Bisexual/Pansexual	263 (53.4)
Lesbian/Gay	115 (23.6)
Queer	64 (13.1)
Unsure/Questioning	21 (4.3)
Asexual	12 (2.5)
Not Listed	7 (1.4)
Straight/Heterosexual	6 (1.2)
Gender Identity
Female	360 (73.8)
Gender Non-Conforming/Gender Queer/Non-Binary	77 (15.8)
Transgender/Male	44 (9.0)
Not Listed	7 (1.4)

Procedure

Participants from FAB 400 and Project Q2 were recruited using venue-based recruitment, social media, and incentivized snowball sampling. For previous publications using data from FAB 400, see (Dyar et al., 2020; Swann et al., 2021; Swann et al., 2020). Participants completed assessments at 6-month intervals (except for the interval between Wave 6 and 7, which was one year) spanning 3.5 years and were compensated $50 for each visit. The study protocol was approved by the Institutional Review Board at Northwestern University with a waiver of parental permission for participants under 18 years of age under 45 CFR 46, 408(c). Participants provided written informed consent, and we used mechanisms to safeguard participant confidentiality (i.e., a federal certificate of confidentiality). The present study used data from Waves 1–7. Retention rates for the cohort at Waves 2–7 were as follows: 96.3%, 94.7%, 92.8%, 92.4%, 90.3%, and 87.9%. For the present analyses, only data from waves at which participants reported having had at least one romantic partner in the last six months were included (analytic N = 488; average number of waves per participant: 5.4).

Measures

Minority Stress

SGM Victimization.

Ten items assessed SGM victimization (Mustanski et al., 2016). At each wave, participants were asked to indicate how often in the past six months each of the following occurred because they are, or were thought to be gay, lesbian, bisexual, or transgender. Sample item: “Have you been verbally insulted (yelled at, criticized), because you are, or were thought to be gay, lesbian, bisexual, or trans*?” Items are rated on a scale from 0 (Never) to 5 (More than ten times). Items are averaged, with higher scores reflecting more frequent experiences of SGM victimization (α = .75). The measure showed evidence of a unidimensional factor structure and predictive validity in a sample of SGM adolescents and young adults (Mustanski et al., 2016).

Sexual Orientation Microaggressions.

At each wave, participants completed the 19 item Sexual Orientation Microaggressions Inventory (SOMI; Swann et al., 2016). Participants were asked “In the past six months, how often have you had the following experiences?” Sample item: “You heard someone say ‘that’s so gay’ in a negative way.” Items are rated on a scale from 1 (Not at all) to 5 (Almost every day) and averaged, with higher scores reflecting more frequent experiences of sexual orientation microaggressions (α = .92). The measure showed convergent, criterion-related, and discriminant validity in two diverse samples of LGBTQ youth (Swann et al., 2016).

Internalized Heterosexism.

Participants completed the Desire to be Heterosexual subscale of Puckett et al. (2017)’s adapted and validated measure at each wave. Participants rated their agreement with each of eight statements (e.g., “Sometimes I think that if I were straight, I would probably be happier”; 1 = strongly disagree to 4 = strongly agree). For items that included reference to participants’ sexual orientation, each participant’s sexual orientation was piped in from responses on the demographic survey. The measure was developed on a sample of sexual and gender minorities assigned male at birth (SGM-AMAB); thus, the word “men” was changed to “women” in several items for our sample. Scores reflect the average response across items, with higher scores indicating higher levels of internalized heterosexism (α = .86). The measure showed good convergent, discriminant, and predictive validity in a sample of sexual minority youth (Puckett et al., 2017).

Intimate Partner Violence

At each visit, five types of IPV (described below) were assessed for up to three partners in the last six months; thus, participants could complete IPV items up to three times (one time for each partner) at each wave. For each type of IPV, items were administered twice: once to assess victimization and once to assess perpetration. On each item on the five IPV scales described below, participants indicated how frequently each behavior or event occurred on a scale ranging from 0 (never) to 6 (more than 20 times). Because participants completed separate measures of IPV for each of up to three relationships in the previous six months, summing scores across multiple partners would not be a meaningful way to reflect frequency of IPV in the previous six months. Thus, we used responses on each scale to create a dichotomous variable indicating occurrence of each of the five types of IPV victimization and perpetration. If participants reported a value greater than zero for any of the items for any partner on a subscale, they were coded as 1; those that reported zero for all items across all partners in the last six months were coded as 0.

Physical, Psychological, and Sexual IPV.

To assess physical, psychological, and sexual IPV, participants completed the Sexual and Gender Minorities Conflict Tactics Scale (SGM-CTS2), a newly developed version of the CTS2 (Straus et al., 1996) adapted to be culturally appropriate for SGM samples (Dyar et al., 2021). In the psychometric evaluation study of the SGM-CTS2, the measure showed the same factor structure of the CTS2 and evidence of validity and reliability (Dyar et al., 2021). For this paper, we used a seven-item version of the SGM-CTS2 assessing psychological (three items), physical (three items) and sexual IPV (one item), because the full version was only administered regarding one partner per visit. Scores on the seven-item version of the SGM-CTS2 showed high correlations with scores on the full scale for perpetration (r = .91) and victimization (r = .92).

Coercive Control.

Coercive control, or behaviors that are intended to monitor and control an intimate partner, was measured using a brief two-item version of an investigator-developed scale based on the Controlling Behaviors Scale (Frankland & Brown, 2014) and items from the National Intimate Partner and Sexual Violence Survey (NISVS; Walters et al., 2013). The items were: “[Partner name] monitored my time and made me account for my whereabouts” and “[Partner name] made it difficult for me to see friends or family.” In the psychometric evaluation study of the Coercive Control measure, the full eight item measure showed the expected unidimensional factor structure and evidence of convergent and divergent validity (Dyar et al., 2021). Scores on the two-item version of the Coercive Control scale showed high correlations with scores on the full scale for perpetration (r = .87) and victimization (r = .89).

Sexual and Gender Minority-Specific IPV.

The two-item SGM-specific IPV Tactics Scale (Dyar et al., 2021) was developed to measure forms of IPV that are unique to SGM individuals, which involve leveraging stigma against SGM to control or cause harm to an intimate partner (Whitton, Dyar, et al., 2019). The items were: “[Partner name] threatened to out me to my friends, family, or other people in my life if I didn’t do what they wanted” and “[Partner name] repeatedly told me how alone I would be if I left them, because other people don’t know about or aren’t accepting of my sexuality.” Scores on the two-item version of the SGM-specific IPV Tactics Scale showed high correlations with scores on the full scale for perpetration (r = .91) and victimization (r = .78).

Data Analysis Plan

Analyses were conducted in Mplus Version 8.6. Monte Carlo simulations in Mplus were used to estimate the sample size required to detect small (OR = 1.2, equivalent to Cohen’s d < .2; Chen et al., 2010) effects of SGM victimization, sexual orientation microaggressions, and internalized heterosexism on IPV at the within- and between-persons levels. Using 500 simulated datasets, we determined that with a sample size of 488 and an average cluster size (i.e., number of completed waves) of 5, we would have over 90% power at a p-value of .05 to detect small effects of SGM victimization, sexual orientation microaggressions, and internalized heterosexism on IPV at the within- and between-persons levels. Our analytic sample included 2,636 observations and 488 participants, for an average cluster size of 5.4, which indicated that we would be sufficiently powered to detect small effects of minority stress on IPV. Within completed assessments, .007% of data were missing and handled using full information maximum likelihood.

We tested associations of minority stress and IPV using multilevel models with a robust maximum likelihood estimator. Predictors were separated into within- and between-persons components by person- and grand-mean centering. Prior to centering, scores on the SGM victimization and microaggressions scales were standardized because their means and standard deviations were small, causing inflated odds ratios. In a separate model for each IPV outcome, we entered all three minority stress variables simultaneously to predict IPV at within- and between-persons levels. To test Hypotheses 1 and 3, in one set of these models, we estimated concurrent within-persons associations (minority stress variables at each wave predicting IPV at the same wave) and between-persons associations (minority stress variables predicting IPV, averaged across waves). To test Hypothesis 2, an additional set of models tested prospective within-persons associations (the three minority stress variables predicting IPV at the next wave, controlling for the outcome at the previous wave). Due to the large number of tests performed, we used the Benjamini and Hochberg (1995) correction in R to adjust p-values.

Age at Wave 1, race/ethnicity, sexual orientation, and gender identity were included as covariates in all models. Race/ethnicity was recoded into four categories: White, Black, Latinx, and Other; White was used as the referent group. Gender identity was recoded into two categories: Cisgender and Gender Minority; Cisgender was used as the referent group. Sexual orientation was recoded into three categories: Lesbian, Bisexual/Queer/Pansexual, and Other; Lesbian was used as the referent group.

Results

Preliminary Results

See Table 2 for correlations, means, standard deviations, and intraclass correlations (ICCs) for predictor variables. In terms of rates of IPV across all partners in the last six months, psychological IPV perpetration was reported at 48.5% of visits, psychological IPV victimization was reported at 45.6% of visits, physical IPV perpetration was reported at 11.8% of visits, and physical IPV victimization was reported 11.5% of visits. Sexual IPV perpetration and victimization were both reported at less than 1% of visits (10 visits and 72 visits, respectively). Coercive control perpetration was reported at 14.7% of visits, and coercive control victimization was reported at 19.4% of visits. Lastly, SGM-specific IPV perpetration was reported at 1.1% of visits, and SGM-specific IPV victimization was reported at 3.8% of visits.

Table 2.

Correlations and Descriptive Statistics for Predictor Variables

Variable	SGM Victimization	Microaggressions	Internalized heterosexism
SGM victimization	—	46**	09**
Microaggressions	.75**	—	17**
Internalized heterosexism	14**	.21**	—
Mean	.14	1.52	1.60
Standard deviation	.31	.51	.55
Range	0–5	1–5	1–4
ICC	.45	.48	.65

Note. ICC = intraclass correlation coefficient. Correlations above the diagonal are within-persons correlations; correlations below the diagonal are between-person correlations.

^*p < .05

^**p < .001

Within-Person Concurrent, Within-Person Prospective, and Between-Persons Associations of Minority Stress and Intimate Partner Violence

SGM Victimization.

Results for tests of Hypotheses 1 and 2 indicated that there were no significant within-persons concurrent or prospective associations of SGM victimization with IPV victimization or perpetration. Regarding results for Hypothesis 3, controlling for the other minority stress variables, SGM victimization showed significant between-persons associations with all types of IPV victimization except for coercive control (see Table 3), and with psychological and sexual IPV perpetration (see Table 4). This indicates that, after accounting for experiences of sexual orientation microaggressions and internalized heterosexism, participants who experienced more acute SGM victimization events on average across waves were more likely to experience psychological, physical, and sexual IPV victimization, and more likely to engage in perpetration of psychological and sexual IPV.

Table 3.

Odds Ratios and 95% Confidence Intervals for Multivariate Models Testing Concurrent and Prospective Associations Between Minority Stress and IPV Victimization

		Type of IPV Victimization
Predictor	Level	Psychological	Physical	Sexual	Coercive Control	SGM-Specific
SGM Victimization	Within Concurrent	1.32 [1.02,1.70]	1.16 [0.95,1.42]	1.11 [0.79,1.55]	1.17 [0.98,1.40]	1.12 [0.91,1.37]
	Within Prospective	1.19 [0.94,1.50]	0.96 [0.78,1.19]	0.83 [0.54,1.27]	1.11 [0.92,1.33]	0.93 [0.64,1.35]
	Between	2.01 [1.30,3.10]	1.77 [1.16,2.69]	1.59 [1.18,2.14]	1.28 [0.92,1.78]	2.24 [1.47,3,42]
Microaggressions	Within Concurrent	1.38 [1.12,1.71]	1.20 [0.95,1.50]	1.34 [0.94,1.91]	1.48 [1.20,1.83]	1.30 [1.00,1.70]
	Within Prospective	1.03 [0.84,1.25]	1.20 [0.94,1.52]	1.34 [0.81,2.23]	1.10 [0.89,1.37]	1.52 [1.01,2.28]
	Between	1.95 [1.33,2.87]	1.52 [1.00,2.31]	1.15 [0.73,1.81]	1.69 [1.17,2.43]	1.31 [0.81,2.12]
Internalized heterosexism	Within Concurrent	1.23 [0.89,1.71]	1.11 [0.71,1.75]	1.87 [0.66,5.33]	1.30 [0.88,1.92]	1.02 [0.42,2.48]
	Within Prospective	1.00 [0.66,1.52]	1.34 [0.75,2.39]	0.70 [0.26,1.90]	1.37 [0.88,2.11]	1.91 [0.78,4.66]
	Between	0.86 [0.55,1.33]	1.11 [0.64,1.97]	1.54 [0.82,2.90]	1.27 [0.77,2.08]	1.65 [0.80,3.41]

Note. IPV = intimate partner violence. SGM = sexual and gender minority. Models included all three minority stress variables predicting each IPV outcome. All models were estimated controlling for age at baseline, race/ethnicity, sexual orientation, and gender identity. Statistically significant associations (p < .05) are in bold text, using p-values adjusted with the Benjamini and Hochberg correction.

Table 4.

Odds Ratios and 95% Confidence Intervals for Multivariate Models Concurrent and Prospective Associations Between Minority Stress and IPV Perpetration

		Type of IPV Perpetration
Predictor	Level	Psychological	Physical	Sexual	Coercive Control	SGM-Specific
SGM Victimization	Within Concurrent	1.27 [0.99,1.64]	1.15 [0.93,1.43]	1.10 [0.83,1.46]	1.29 [1.05,1.59]	1.65 [1.10,2.48]
	Within Prospective	1.16 [0.95,1.43]	0.96 [0.73,1.25]	0.05 [0.00,3.37]	0.95 [0.74, 1.21]	0.88 [0.43,1.80]
	Between	2.23 [1.36,3.68]	1.43 [0.86,2.36]	2.21 [1.25,3.92]	0.86 [0.62,1.18]	1.36 [0.72,2.58]
Microaggressions	Within Concurrent	1.36 [1.10,1.68]	1.43 [1.14,1.78]	2.39 [1.38,4.15]	1.34 [1.06,1.68]	0.97 [0.54,1.76]
	Within Prospective	1.11 [0.91,1.35]	1.14 [0.88,1.47]	2.22 [0.50,9.97]	1.37 [1.06,1.77]	6.92 [2.11,22.68]
	Between	1.75 [1.14,2.69]	1.66 [1.05,2.63]	0.94 [0.40, 2.20]	2.15 [1.39,3.33]	1.51 [0.61,3.74]
Internalized heterosexism	Within Concurrent	1.54 [1.07,2.22]	0.96 [0.59,1.57]	0.22 [0.03,1.71]	1.64 [1.05,2.57]	1.20 [0.32,4.57]
	Within Prospective	0.99 [0.67,1.48]	0.91 [0.51,1.64]	1.17 [0.24,5.69]	1.79 [1.10,2.92]	0.10 [0.01,1.22]
	Between	0.97 [0.60,1.56]	0.92 [0.60,1.70]	0.62 [0.16,2.38]	1.77 [1.04,3.03]	2.38 [0.73,7.82]

Note. IPV = intimate partner violence. SGM = sexual and gender minority. Models included all three minority stress variables predicting each IPV outcome. All models were estimated controlling for age at baseline, race/ethnicity, sexual orientation, and gender identity. Statistically significant associations (p < .05) are in bold text, using p-values adjusted with the Benjamini and Hochberg correction.

Sexual Orientation Microaggressions.

Results for Hypothesis 1 showed that, controlling for SGM victimization and internalized heterosexism, microaggressions had significant concurrent within-persons associations with psychological IPV and coercive control victimization (see Table 3) and psychological, physical, and sexual IPV perpetration (see Table 4). This indicates that, within individuals, after accounting for SGM victimization and internalized heterosexism, risk for experiencing psychological IPV and coercive control victimization, and engaging in psychological, physical, and sexual IPV perpetration, was higher at waves when individuals experienced more microaggressions than at other waves. For Hypothesis 2, microaggressions also had a significant prospective within-persons association with perpetration of SGM-specific IPV (see Table 4). When an individual experienced more microaggressions than usual at a given wave, they experienced increased odds of SGM-specific IPV perpetration during the following 6 months. Lastly, regarding Hypothesis 3, between-persons, controlling for SGM victimization and internalized heterosexism, microaggressions was significantly associated with psychological IPV and coercive control victimization (see Table 3), and with perpetration of coercive control (see Table 4). These results indicate that, after accounting for experiences of SGM victimization and internalized heterosexism, participants who experienced more frequent microaggressions on average across waves were more likely to experience psychological IPV and coercive control victimization, and to engage in perpetration of coercive control.

Internalized Heterosexism.

Results showed that Hypotheses 1 and 2 were not supported: Internalized heterosexism showed no significant concurrent or prospective within-persons associations with any type of IPV victimization or perpetration. Hypothesis 3 was also unsupported, as internalized heterosexism was unassociated with any IPV outcome at the between-persons level.

Sensitivity Analyses

Because of the 12-month interval between Waves 6 and 7, we conducted sensitivity analyses with data from Wave 7 removed from models testing prospective associations of minority stress with IPV. We found no differences in prospective associations.

To account for possible differences in results because of the inclusion of data from participants who were not retained in all study waves, we conducted additional sensitivity analyses with data from participants who were not retained across all seven waves (N = 144) removed from the analytic sample. All associations that were positive and significant in the full sample were in the same direction; however, only the between-persons associations of microaggressions with psychological IPV victimization and coercive control perpetration remained significant.

We also tested for differences on demographic and baseline variables between participants with complete data and those who missed at least one study wave. An independent-samples t-test showed that participants who completed all study waves were significantly younger (M = 19.67, SD = 4.32) than participants who missed at least one study wave (M = 21.48, SD = 3.37), t(486) = 4.55, p < .001. Chi-squared tests of independence showed that, compared to participants who completed all study waves, those who missed at least one wave were more likely to be Black, X² (1, N = 488) = 13.39, p < .001, and cisgender X² (1, N = 488) = 6.67, p = .010.

We also found that those who missed at least one study wave had significantly higher Wave 1 scores on SGM victimization (M = 0.70, SD = 2.17) compared to those who completed all study waves (M = 0.27, SD = 1.19), t(486) = 2.64, p < .001 and higher Wave 1 scores for sexual orientation microaggressions (M = 0.79, SD = 1.57) compared to those who completed all waves (M = 0.43, SD = 1.16), t(480) = 2.53, p < .001. Lastly, those who missed at least one study wave were more likely to report having experienced psychological IPV perpetration X² (1, N = 407) = 5.22, p = .022, and victimization X² (1, N = 407) = 5.22, p = .022, sexual IPV perpetration X² (1, N = 405) = 6.88, p = .009, coercive control victimization X² (1, N = 407) = 5.65, p = .017, and SGM-specific IPV perpetration X² (1, N = 407) = 4.86, p = .028 at Wave 1, compared to participants who completed all study waves.

Discussion

The present study used data from multiple waves of a longitudinal cohort study of young SGM-AFAB to investigate within- and between-persons associations of minority stress with IPV victimization and perpetration. In support of Hypothesis 1, our results indicated that there were significant within-persons concurrent associations of sexual orientation microaggressions and IPV. Specifically, we found that after accounting for experiences of SGM victimization and internalized heterosexism, at waves when SGM-AFAB experienced more microaggressions than usual, they were more likely to experience psychological IPV and coercive control victimization and perpetrate psychological, physical, and sexual IPV. Previous studies on distal minority stressors and IPV have also shown that experiences of discrimination have been associated with past year physical, psychological, and sexual IPV perpetration and victimization among SGM-AFAB (Balsam & Szymanski, 2005; Do et al., 2022). The present study expands on previous cross-sectional studies by showing that, within individuals, at times when experiences of microaggressions are more frequent than usual, there were also increases in likelihood of several types of IPV victimization and perpetration.

However, Hypothesis 1 was only partially supported, because the other minority stress predictors we examined were not associated with IPV at the within-persons level. Hypothesis 3 was also only partially supported, showing a similar pattern of results. Participants who, on average across study waves, experienced more distal minority stressors (i.e., SGM victimization and sexual orientation microaggressions) were more likely to experience certain types of IPV victimization and perpetration, but after accounting for the experiences of those stressors, internalized heterosexism was not associated with IPV. Specifically, participants with higher average levels of SGM victimization experiences also had increased likelihood of experiencing psychological, physical, and sexual IPV victimization, as well as psychological and sexual perpetration, across study waves. In addition, SGM-AFAB who experienced higher average levels of sexual orientation microaggressions were more likely to experience psychological IPV and coercive control victimization and were more likely to perpetrate coercive control.

Taken together, these results appear to indicate that both concurrently within-persons and between-persons, distal minority stressors are more salient potential risk factors for IPV among SGM-AFAB in relationships, and that internalized heterosexism may not have a unique impact on IPV after taking into account the impact of experiences of SGM victimization and sexual orientation microaggressions. This pattern of findings is somewhat consistent with those observed by Balsam and Szymanski (2005), who included discrimination and internalized heterosexism in the same model and found that discrimination was a stronger predictor of IPV than internalized heterosexism; although, in contrast to our results, Balsam and Szymanski (2005) found that internalized heterosexism was still significantly associated with IPV. Notably, the other two studies that found associations of internalized heterosexism with IPV perpetration and victimization among women in same-sex relationships examined only internalized heterosexism and did not include distal minority stressors (Lewis et al., 2014; Milletich et al., 2014), making it less clear how their results would have been consistent or inconsistent with the results of the present study.

In addition, although both types of distal minority stressors were associated with IPV, sexual orientation microaggressions was the only predictor associated with IPV within-persons, indicating that as microaggressions fluctuated from wave to wave, likelihood of experiencing IPV did as well. It is notable that mean frequency of experiences of microaggressions was higher than that of SGM victimization in our sample; this higher frequency, combined with an ICC of .48, indicating that 52% of the variance in microaggressions was due to within-person fluctuations from wave to wave, may explain why concurrent within-persons associations were only found for this variable in the present study. In contrast, although SGM victimization had a similar proportion of variation from wave to wave, its lower frequency may have meant that it was less impactful on IPV as it fluctuated from wave to wave.

Hypothesis 2 was partially supported, in that we found only one within-persons prospective association of minority stress with IPV. In other words, for SGM-FAB participants, experiencing more minority stress than usual at one wave was not as typically associated with increased likelihood of IPV six months later, although we did find one significant prospective association of microaggressions with SGM-specific IPV perpetration. This pattern of findings may indicate that experiences of minority stress increase the likelihood of IPV behaviors in the short-term, but do not typically exert long-term effects on IPV.

The current study contributes novel information to the field by examining SGM-specific IPV, a unique and under-studied form of IPV related to the marginalized status of SGM communities (e.g., threatening to “out” one’s partner at work; Burke et al., 2002). In addition to finding that at waves when participants experienced more microaggressions, they were more likely to perpetrate SGM-specific IPV at the following wave six months later, we found that participants who experienced more SGM victimization, on average, were more likely to experience SGM-specific IPV victimization. These results contrast with those of Balsam and Szymanski (2005), who found no significant associations of discrimination or internalized heterosexism with SGM-specific IPV.

Limitations

Results of this study should be considered in light of several limitations. First, the six-month lag between study visits could have limited our ability to detect within-persons prospective associations of minority stress with IPV. Second, our recruitment methods may limit the generalizability of our results to those who are more out and connected to the SGM community than other SGM-AFAB, which can impact rates of both minority stress and IPV. The generalizability of our results may also have been impacted by differences between participants with complete data and those who missed at least one study wave, which indicated that those with complete data had significantly lower mean scores on distal minority stress and certain IPV outcome variables at Wave 1. Third, although the total sample size was substantial for a longitudinal cohort study of SGM-AFAB, numbers of specific gender, sexual orientation, and racial/ethnic subgroups were too small to explore potential demographic differences in associations of minority stress and IPV, nor did we examine unique forms of stigma that are relevant to subgroups of SGM-AFAB (e.g., sexism, cissexism, monosexism, or racism). Fourth, to reduce participant burden, we administered abbreviated measures of IPV, some of which included one or two items, which imposes limitations on their psychometric properties. Additional evaluation of the psychometrics of these measures is needed. Lastly, because we used dichotomous IPV outcome variables to best capture experiences across multiple partners, we were not able to provide evidence of how minority stress is associated with frequency or severity of IPV perpetration and victimization.

Future Research Directions

To address the limitations of the present study, additional research using shorter intervals between assessments, such as daily diaries or experience sampling studies, may provide more a fine-grained examination of prospective associations of minority stress with IPV. Future research that includes more representative samples of SGM-AFAB (e.g., those who are less out or connected to SGM communities, those with more frequent distal minority stress and IPV experiences), and with samples that include larger gender, sexual orientation, and racial/ethnic subgroups would increase the generalizability of results and allow for the examination of potential demographic differences in results. In addition, because research on SGM-specific IPV and its correlates is still in its infancy, an important future research direction is to replicate the results of the present study and further explore its antecedents and consequences. Further research is also needed on the mechanisms of associations of minority stress and IPV. For example, chronic experiences of stigma can increase the likelihood of mental health disorders and maladaptive forms of coping (Do et al., 2022; Lewis et al., 2015), negatively impact relationship quality (Balsam & Szymanski, 2005), and increase negative affect and intrusiveness (i.e., a lack of boundaries and autonomy in a relationship; Mason et al., 2016), all of which can increase risk for IPV. Elucidating these potential mediators is an important direction for future research and to further inform interventions to reduce prevalence of IPV among SGM-AFAB

Clinical Implications

Results of this study have important clinical implications. For clinicians working with SGM-AFAB in relationships, screening for minority stress experiences, particularly SGM victimization and microaggressions, may help identify those at risk for IPV. Clients who report elevated experiences of minority stress may benefit from a thorough assessment for IPV, and interventions to reduce minority stress may help prevent IPV. This can be achieved through broad antidiscrimination policies to reduce the stigmatization of SGM at a societal level, as well as individual-level interventions that incorporate strategies to help SGM cope with minority stress. Interventions to enhance strategies for coping with experiences of minority stress to reduce the occurrence of potential mediating factors (e.g., mental health disorders, maladaptive forms of coping, decreased relationship quality, increased negative affect and intrusiveness) may help to prevent IPV among SGM-AFAB.