Stigma, Social Cohesion, and Mental Health Among Transgender Women Sex Workers Living with HIV in the Dominican Republic

Abstract

Globally, transgender (trans) women experience pervasive stigma, which contributes to health inequities across multiple mental health outcomes. While trans stigma has been associated with increased mental health challenges, less is known about how stigma related to sex work and HIV affect mental health. Social cohesion may ameliorate the negative health consequences of stigma, but little is known about how social cohesion is associated with mental health among trans women. Using cross-sectional survey data collected in 2019 among 100 trans women sex workers living with HIV in Santo Domingo, Dominican Republic, we examined associations between stigma (related to HIV, sex work, and trans identity), social cohesion, and depression and anxiety symptoms. In unadjusted models, enacted and internalized sex work stigma were associated with greater odds of depression and anxiety symptoms and social cohesion was associated with less depression and anxiety. Resisted sex work stigma, which captured acceptance of self and work, was associated with lower odds of depression and trans stigma was associated with increased anxiety. In adjusted models, internalized sex work stigma was associated with increased depression (aOR=1.19, p=0.018) and anxiety symptoms (aOR=1.12, p=0.021), while social cohesion was associated with reduced depression (OR=0.81, p=0.002) and anxiety symptoms (aOR=0.086, p=0.023). Findings indicate the need for interventions to focus on reducing internalized sex work stigma and promoting social cohesion to improve mental health among trans women. Additional research is needed to determine how to best measure the intersections between HIV, sex work, and trans stigma and their impact on mental health.

INTRODUCTION

Globally, transgender (trans) women experience inequities across multiple mental health outcomes, including depression and anxiety. In one study among a sample of nearly 28,000 trans and other gender diverse people in the US, 39.0% of respondents reported experiencing psychological distress in the past month, compared with 5.0% among a national sample of cisgender people (James et al., 2016). Trans women also experience an increased odds of HIV, with global HIV prevalence among trans women estimated to be approximately 19.0% (Baral et al., 2013). Due to stigma, sex work is often the primary employment option for trans women (Barrington, Acevedo, Donastorg, Perez, & Kerrigan, 2017; Nadal, Davidoff, & Fujii-Doe, 2014); HIV inequities are heightened for trans women sex workers, with a global HIV prevalence of approximately 27.0%, much higher than the global estimate of 10.8% among cisgender women sex workers (Baral et al., 2012; Operario, Soma, & Underhill, 2008).

HIV and mental health are reciprocally related to each other; people living with HIV experience disproportionate burdens of depression and anxiety, starting with diagnosis and throughout managing care and treatment (Collins, Holman, Freeman, & Patel, 2006; Remien et al., 2019). Marginalized populations may also experience structural vulnerabilities and stressors (e.g., poverty, homelessness) that may simultaneously contribute to risks for HIV and poor mental health (Remien et al., 2019). At the same time, mental health conditions can contribute to poor HIV outcomes, such as lower retention in HIV care and lower adherence to antiretroviral therapy (Collins et al., 2006; Remien et al., 2019; Rooks-Peck et al., 2018). These findings highlight the importance of understanding and promoting mental health among people living with HIV; however, little is known about the mental health of trans women living with HIV.

Globally, trans populations experience pervasive stigma, which can limit access to resources, contribute to systemic vulnerability (e.g., housing insecurity, unemployment, poverty), and generate chronic stress (Hatzenbuehler, Phelan, & Link, 2013; Link & Phelan, 2001; Meyer, 2003; Reisner et al., 2016; Testa, Habarth, Peta, Balsam, & Bockting, 2015). Stigma can simultaneously increase the risk of experiencing poor mental health, while also reducing access to services that may promote mental health (Hatzenbuehler et al., 2013). Prior research demonstrates that enacted, anticipated, and internalized trans stigma is associated with increased depression, anxiety, post-traumatic stress, and suicide (Bockting, Miner, Swinburne Romine, Hamilton, & Coleman, 2013; Nuttbrock et al., 2014; Testa et al., 2015). In fact, historically, trans stigma has been rooted within the context of mental health as trans identity was classified as a mental health disorder, which perpetuated both trans stigma and mental health stigma, and created mistrust of mental health services, which may further exacerbate mental health inequities experienced by trans people (Dewey & Gesbeck, 2017; Hughto, Reisner, & Pachankis, 2015).

Less is known about the associations between other forms of stigma, in particular stigma related to HIV and sex work, and mental health outcomes among trans women. An intersectional stigma framework recognizes that individuals may experience multiple types of stigma simultaneously and that these different types of stigma may be interrelated (Turan et al., 2019). For example, high HIV prevalence among trans women can result in HIV stigma (especially anticipated HIV stigma) for all trans women, regardless of their HIV status (Golub & Gamarel, 2013), and can fundamentally shape experiences related to both HIV and trans identity. While there are methodological challenges to analyzing and interpreting the health effects of intersecting types of stigma (Turan et al 2019), in this paper we aim to measure and assess whether stigma related to HIV, sex work and trans identity are associated with mental health outcomes among trans women sex workers living with HIV.

An intersectional framework also considers how having multiple marginalized identities can shape the ways in which individuals or groups resist multiple forms of stigma and experience resilience (Turan et al., 2019). Social cohesion refers to the connectedness, solidarity, and mutual support between members of a community or group. Trans women in Peru use social cohesion as a mechanism for managing stigma and HIV vulnerability (Perez-Brumer et al., 2017), and global research with cisgender women sex workers has found associations between social cohesion and protective HIV-related behaviors (e.g., condom use) (Carrasco, Barrington, Donastorg, Perez, & Kerrigan, 2014; Carrasco et al., 2017; Fonner et al., 2014; Kerrigan, Telles, Torres, Overs, & Castle, 2008; Lippman et al., 2012; Lippman et al., 2010). Social cohesion may also promote mental health through an increased sense of belonging with a community, increased access to community resources (e.g., through an ability to engage in community-based organizations), and the development of alternative societal values and norms that more appropriately apply to the perspectives and experiences of marginalized populations (Carrasco et al., 2017; Frost & Meyer, 2012; Meyer, 2015). However, social cohesion and trust have also been found to be low among trans women in diverse settings (Barrington et al., 2020; Kaplan et al., 2015). Little is known about how social cohesion is experienced among trans women sex workers living with HIV or the associations between social cohesion and mental health in this population. Therefore, the purpose of this paper is to improve understanding of the associations between stigma related to HIV, sex work and trans identity and social cohesion and mental health outcomes among trans women sex workers (who report exchanging sex for money) living with HIV.

METHODS

Study setting.

This study took place in Santo Domingo, the capital of the Dominican Republic (DR). Adult trans women account for approximately 0.2% of the Dominican population (Edwards et al., 2018). Research demonstrates that trans women in the DR experience a disproportionate burden of poor health outcomes, including HIV and mental health conditions (Barrington et al., 2017; Budhwani et al., 2018; Budhwani, Hearld, Milner, et al., 2017; Kerrigan et al., 2006; Milner et al., 2019). Trans women in the DR also experience pervasive trans stigma and discrimination across multiple settings, including, for example, in employment, education, and health care (Barrington et al., 2017; Budhwani, Hearld, Hasbun, et al., 2017; Budhwani et al., 2018; Budhwani, Hearld, Milner, et al., 2017; Milner et al., 2019). As a result, trans women in the DR identify sex work and beauty salons as their main employment opportunities (Barrington et al., 2017). Trans women sex workers living with HIV have also reported low social cohesion and high levels of mistrust with other trans women sex workers (Barrington et al., 2020). Notably, trans women in the DR who engage in sex work report experiencing more stigma, less social support, more violence, and a lower quality of life when compared with trans women in the DR who are not engaging in sex work (Milner et al., 2019).

Study procedures and sample.

Interviewer-administered surveys were conducted from March to August 2019 in Spanish. Interviewers (all cisgender women) were research staff at the Instituto Dermatológico y Cirugía de Piel (IDCP) in Santo Domingo, DR who were all psychologists trained in procedures for maintaining confidentiality. IDCP has a longstanding history of working with trans people, sex workers, and people living with HIV. Interviews took place in a private space at IDCP and were conducted using a tablet device. All participants provided verbal informed consent and received $10 (paid in DR pesos) to compensate them for their time. A referral list for support services (including mental health and substance use services) was available for all participants. All study activities were approved by Institutional Review Boards at the Johns Hopkins Bloomberg School of Public Health in the United States and the IDCP and Consejo Nacional de Bioteca (CONABIOS) in the DR.

Eligibility criteria included being at least 18 years of age, being assigned a male sex at birth and identifying as a trans woman (which was defined by using a range of locally appropriate terms in Spanish), reporting exchanging sex for money in the past month, and having a confirmed HIV diagnosis using a single rapid test. Participants were recruited through key informant referral, self-referral, and with the help of local navigators, who recruited participants through HIV clinics; navigators also re-engaged trans women who had participated in past research.

Measures.

Dependent variables included measures of mental health and independent variables included measures of stigma, social cohesion, and additional covariates.

Mental health outcomes.

This analysis examined two mental health outcomes: depression and anxiety. Depressive symptoms were measured using the depression module of the Patient Health Questionnaire (PHQ-9), a 9-item scale measuring the severity of depressive symptoms in the past two weeks (Kroenke, Spitzer, & Williams, 2001). For this analysis, depressive symptoms were measured as a binary variable based on a commonly used cutoff (Kroenke et al., 2001) measuring whether participants reported having moderate to severe depression (including moderate, moderately severe, and severe depression) or less than moderate depression (including no depression, minimal depression, and mild depression). Anxiety symptoms were measured using the anxiety module of the Hospital Anxiety and Depression Scale (HADS-A), a 7-item scale measuring anxiety symptoms in the past two weeks (Spinhoven et al., 1997). For each item, participants were asked about the frequency of symptoms, with response options ranging from never (0) to most of the time (3) for each symptom. For this analysis, anxiety was measured as a binary variable based on a commonly used cutoff (Spinhoven et al., 1997) measuring whether participants reported having no anxiety symptoms (a score of ≤7 on the scale) or any anxiety symptoms (a score of >7 on the scale, which includes both borderline anxiety and anxiety). Measuring depressive symptoms and anxiety symptoms as binary variables, based on well-established cutoffs, allows for meaningful interpretation of results.

Stigma.

Multiple types of stigma were measured, including HIV stigma, sex work stigma, and trans stigma.

HIV stigma. Three separate scales measured internalized, anticipated, and enacted HIV stigma. These scales have been previously used to measure HIV stigma among cisgender women sex workers living with HIV in the DR (Carrasco et al., 2018; Donastorg, Barrington, Perez, & Kerrigan, 2014; Zulliger et al., 2015). The anticipated HIV stigma scale was adapted from Zelaya and colleagues (Zelaya et al., 2012), and is a five-item scale that measured the expectation that HIV stigma would occur. The enacted and internalized HIV stigma scales were both adapted from the short form version of the HIV stigma scale by Berger and colleagues (Berger, Ferrans, & Lashley, 2001). The enacted HIV scale included five items that measured experiences of discrimination and rejection related to HIV status. The internalized HIV scale included ten items to measure participants’ perceptions of themselves and their HIV status. All HIV stigma scales were measured using a 4-point Likert scale ranging from strongly agree to strongly disagree and all demonstrated high internal reliability (Cronbach Alphas: Internalized stigma=0.86, Anticipated stigma=0.80, Enacted stigma=0.71).

Sex work stigma was measured using the Experiences of Sex Work Stigma (ESWS) scale, developed by Kerrigan and colleagues (2021). Scales were developed using Item Response Theory and were based on qualitative research examining the experiences of cisgender women sex workers living with HIV in the DR and in Tanzania (Kerrigan, Karver, Barrington, Davis, et al., 2021). For this analysis, scores were calculated using Differential Item Functioning (DIF) analysis (Kerrigan, Karver, Barrington, Davis, et al., 2021). The DIF analysis used likelihood ratio difference tests to determine whether scale items functioned differently in DR and Tanzania, with tests examining differences in discrimination and location parameters by country; final scores were adjusted to take differences across countries into account (Kerrigan, Karver, Barrington, Davis, et al., 2021). For the current analysis, DIF-adjusted scores were based on the responses of the trans women sex workers in this sample in the DR, compared with the cisgender women sex workers in the DR sample initially used to calculate the scale scores.

The ESWS scale measured four domains of sex work stigma: anticipated stigma (silence scale), enacted stigma (treatment scale), internalized stigma (shame scale), and resisted stigma (dignity scale) (Kerrigan, Karver, Barrington, Davis, et al., 2021). The silence scale was comprised of six items measuring participants’ concealment of sex work based on the expectation that stigma will occur; for example, items include: “You have tried to make sure that no one knows that you do sex work” and “You have done everything you can to keep sex work a secret.” The treatment scale was a nine-item scale that measured enacted sex work stigma, including experiences of rejection and mistreatment due to sex work. Participants indicated how often they experienced the nine items because of their sex work; example items include: being excluded from groups and being called names. The shame scale included six items examining internalized stigma. For this scale, participants indicated their agreement to how often they experienced specific feelings related to their sex work, including for example, feeling ashamed and humiliated. Finally, the dignity scale was comprised of six items and measured positive feelings (e.g., pride, value) representing self-acceptance and acceptance of sex work. All items were measured based on frequency (1=never, 2=sometimes, 3=always) and demonstrated high internal reliability (Cronbach Alphas: Silence=0.91, Treatment=0.82, Shame=0.77, Dignity=0.89).

Trans stigma was measured with a single scale, adapted from Chakrapani and colleagues (Chakrapani et al., 2017). The Transgender Identity Stigma Scale, developed by Chakrapani and colleagues (2017), measures exposure to trans stigma; this scale was originally tested within the context of trans women’s experiences in India (Chakrapani et al., 2017), and was adapted to include items with cultural relevance and salience in the DR context. The adapted scale included 11 items measuring multiple aspects of stigma, including enacted stigma (e.g., discrimination, victimization, and rejection), anticipated stigma, and identity concealment (Cronbach’s Alpha=0.81).

Social Cohesion was measured with an adapted scale (Kerrigan et al., 2008; Lippman et al., 2010) that has previously been used to measure social cohesion among sex workers in the DR (Carrasco et al., 2018). The scale was comprised of 11 items using a 4-point Likert scale (ranging from strongly disagree to strongly agree) to assess the relationships and community support and connectedness that participants had with other trans women sex workers (e.g., trusting other trans women sex workers, relying on other trans women sex workers). The scale demonstrated high internal reliability, with a Cronbach alpha of 0.83.

Covariates included social demographics (age and education) and substance use. Age was measured in years as a continuous variable. Education was measured as a binary variable based on whether participants had attained some high school education. Substance use, including alcohol and drug use, was examined as an important syndemic health outcome among trans women (Brennan et al., 2012; Operario & Nemoto, 2010). Alcohol use was a binary variable based on whether participants reported at least one incident of binge drinking weekly. Drug use was also included as a binary variable, defined as the use of any illicit drugs in the past six months; participants were asked separate questions about their use of marijuana, crack, cocaine, heroin, ecstasy, and other drug use in the past six months and were considered to have used drugs if they indicated having used any of these drugs. Marijuana was included as an illicit drug because it is illegal in the DR and is considered taboo.

Analysis.

Data were analyzed using Stata 14 (College Station, Texas). Unadjusted and adjusted logistic regression models examined associations between each stigma variable and each mental health outcome. There were no missing data, so all 100 study participants were included in the analysis. None of the independent variables demonstrated multicollinearity. Descriptive statistics assessed the sample distributions of all dependent and independent variables.

Unadjusted logistic regression models were fit to examine the independent relationships between stigma variables (HIV, sex work, trans identity) and mental health outcomes (depressive symptoms and anxiety). Then, adjusted models were fit, including only the variables that were significantly associated with the outcome in unadjusted models and covariates (i.e., demographics and substance use). This analytic approach was used as a first step in determining how the distinct types of stigma are associated with mental health outcomes. While this analytic approach did not allow for a nuanced assessment of the intersections of these distinct types of stigma and social cohesion, focusing on the direct associations between each type of stigma and social cohesion aided in reducing the number of variables in each model, which helped to increase the power to detect significant associations as a first step in understanding these complex relationships (Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996). Additionally, as noted above, the intersectional nature of stigma may be captured in the stigma-specific measure given that how trans women experience HIV stigma, for example, is related to their identity as trans women. An alpha level of 0.05 was used to determine significance for all analyses.

RESULTS

Descriptive statistics are presented in Table 1. The median age was 33.5 years and nearly three-quarters of participants had some high school education. Approximately one-quarter (24%) of participants reported experiencing moderate to severe depression in the past two weeks and approximately one-third (34%) reported experiencing anxiety in the past two weeks. Approximately 50% of participants reported binge drinking on a weekly basis and approximately 50% reported drug use within the past six months.

Table 1:

Descriptive Statistics for Mental Health Outcomes (n=100)

Mental health conditions, % (n)
Moderate to severe depression	24.00 (24)
Anxiety	34.00 (34)
HIV stigma, Median (Range)
Internalized	22.00 (10.00–29.00)
Anticipated	14.00 (5.00–20.00)
Enacted	10.00 (5.00–15.00)
Sex work stigma, Median (Range)
Silence (anticipated)	44.25 (29.47–61.24)
Treatment (enacted)	61.35 (47.60–70.45)
Shame (internalized)	59.16 (44.13–79.43)
Dignity (resistance)	48.52 (30.31–62.96)
Trans stigma, Median (Range)	29.00 (11.00–44.00)
Social cohesion, Median (Range)	28.00 (13.00–33.00)
Social Demographics
Age, Median (Range)	33.50 (18.00–60.00)
Graduated high school, % (n)	74.0 (74)
Syndemic health outcomes
Binge drinks ≥ weekly, % (n)	53.0 (53)
Drug use in past 6 months, % (n)	55.0 (55)

Stigma and Social Cohesion.

For HIV stigma, participants had a median score of 22.0 for internalized stigma (range 10.0 to 29.0) on a scale with a possible range of 10–40. The median anticipated HIV stigma score was 14.0 and the median enacted HIV score was 10.0; both of these scales had a range of 5–20, which is the same as the total possible range for the scales. For sex work stigma, participants scored a median of 44.3 for the silence scale (range 29.5 to 61.2), 61.4 for the treatment scale (range 47.6 to 70.5), 59.2 for the shame scale (range 44.1 to 79.4), and 48.5 for the dignity score (range 30.3 to 63.0). For the trans stigma scale, the median score was 29.0, with a range of 11–44 (the same as the total possible range for this scale); 99 of the 100 participants indicated that they had experienced at least one incident of trans stigma at some point in their lives. Finally, the median score on the social cohesion scale was 28.0, with a range of 13–33 on a scale with a possible range of 11–44.

Depression.

In unadjusted models (Table 2), sex work stigma (including enacted sex work stigma, internalized sex work stigma, and resisted sex work stigma) and social cohesion were significantly associated with experiencing moderate to severe depression. Enacted and internalized sex work stigma variables were both associated with greater odds of experiencing depressive symptoms (Enacted: OR=1.28, p=0.002; Internalized: OR=1.18, p<0.001). Resisted sex work stigma (OR=0.95, p=0.049) and social cohesion (OR=0.81, p=0.002) were both associated with lower odds of depressive symptoms. No other variables were significantly associated with depressive symptoms in unadjusted models.

Table 2:

Bivariable and Multivariable Analyses Examining Associations Between Stigma, Social Cohesion, and Mental Health (n=100)

	Moderate to Severe Depression				Anxiety
	Unadjusted OR (95% CI)	p-value	aOR (95% CI)	p-value	Unadjusted OR (95% CI)	p-value	aOR (95% CI)	p-value
HIV stigma
Internalized	1.12 (0.97, 1.29)	0.136	--	--	1.10 (0.97, 1.25)	0.147	--	--
Anticipated	1.07 (0.87, 1.30)	0.532	--	--	1.11 (0.92, 1.33)	0.281	--	--
Enacted	1.22 (0.93, 1.60)	0.155	--	--	1.06 (0.84, 1.33)	0.628	--	--
Sex work stigma
Silence (anticipated)	1.04 (0.99, 1.09)	0.119	--	--	0.98 (0.94, 1.02)	0.317	--	--
Treatment (enacted)	1.28 (1.10, 1.49)	0.002	1.05 (0.87, 1.26)	0.613	1.23 (1.09, 1.39)	0.001	1.05 (0.89, 1.24)	0.556
Shame (internalized)	1.18 (1.08, 1.28)	<0.001	1.19 (1.03, 1.36)	0.018	1.14 (1.06, 1.22)	<0.001	1.12 (1.02, 1.23)	0.021
Dignity (resistance)	0.95 (0.89, 1.00)1	0.049	1.04 (0.94, 1.14)	0.475	0.97 (0.92, 1.01)	0.163	--	--
Transgender stigma	1.06 (1.00, 1.13)	0.063	--	--	1.09 (1.02, 1.16)	0.007	1.01 (0.92, 1.10)	0.910
Social cohesion	0.81 (0.72, 0.93)	0.002	0.82 (0.71, 0.95)	0.009	0.84 (0.74, 0.94)	0.003	0.86 (0.75, 0.98)	0.023
Covariates
Age	--	--	0.99 (0.94, 1.05)	0.817	--	--	1.00 (0.95, 1.06)	0.943
Graduated high school	--	--	1.23 (0.33, 4.54)	0.753	--	--	0.55 (0.17, 1.75)	0.310
Binge drinks ≥ weekly	--	--	1.60 (0.45, 5.71)	0.465	--	--	1.06 (0.36, 3.16)	0.912
Drug use in past 6 months	--	--	2.37 (0.64, 8.71)	0.193	--	--	3.47 (1.11, 10.89)	0.033

¹The 95% CI appears to include 1.00 because of rounding

In the adjusted model, internalized sex work stigma and social cohesion were statistically associated with depression. For each additional point on the internalized sex work stigma scale, the odds of experiencing depressive symptoms increased by 19% (p=0.018), and for each additional point on the social cohesion scale, the odds of reporting depressive symptoms decreased by 18% (p=0.009). None of the covariates were significantly associated with depression.

Anxiety.

In unadjusted models examining anxiety, sex work stigma (including enacted and internalized sex work stigma), trans stigma, and social cohesion were all statistically significant (Table 2). All stigma variables were associated with increased odds of experiencing anxiety symptoms (Enacted sex work stigma: OR=1.23, p=0.001; Internalized sex work stigma: OR=1.14, p<0.001; Trans stigma: OR=1.09, p=0.007), and social cohesion was associated with lower odds of experiencing anxiety symptoms (OR=0.84, p=0.003).

In the adjusted model, internalized sex work stigma, social cohesion, and drug use were statistically significant. For each additional point on the internalized sex work stigma scale, odds of experiencing anxiety symptoms increased by 12% (p=0.021) and for each additional point on the social cohesion scale, odds of experiencing anxiety symptoms decreased by 14% (p=0.023). Participants who reported drug use had 3.47 times greater odds of experiencing anxiety symptoms (p=0.033).

DISCUSSION

This study is the first to examine associations between multiple types of stigma related to HIV, sex work, and trans identity, social cohesion, and mental health among trans women sex workers living with HIV. In unadjusted models, enacted and internalized sex work stigma were significantly associated with an increase in depression and anxiety symptoms and social cohesion was significantly associated with a decrease in depression and anxiety symptoms. Resisted sex work stigma was significantly associated with fewer depressive symptoms and trans stigma was significantly associated with having more anxiety symptoms. In adjusted models, findings were consistent across depression and anxiety, with internalized sex work stigma and social cohesion being significantly associated with both mental health outcomes. Trans stigma was especially high, with only one participant indicating that they had never experienced trans stigma; this is consistent with previous research in the DR that demonstrates that trans stigma may be especially high among trans women in the DR who engage in sex work (Milner et al., 2019).

Findings from the current study add to our knowledge of mental health experiences among a trans population in a resource-poor setting; this is especially important, since most mental health research with trans populations has occurred within the United States, or other high-income countries (Reisner et al., 2016). The prevalence of anxiety symptoms among trans women in this sample is similar to the prevalence reported by cisgender women sex workers living with HIV in the DR (Kerrigan, Karver, Barrington, Donastorg, et al., 2021). However, the prevalence of experiencing moderate to severe depressive symptoms is slightly lower among this sample of trans women (reported at 24%) than among cisgender women sex workers living with HIV in the DR (38%) (Kerrigan, Karver, Barrington, Donastorg, et al., 2021). It is possible that trans women may be more reluctant than cisgender women to report symptoms of depression due to mental health stigma, since trans identity has historically been understood as a mental illness (Dewey & Gesbeck, 2017; Hughto et al., 2015). However, more research exploring the prevalence of mental health conditions among trans populations living in low- and middle-income countries (LMICs) is needed.

The association between internalized sex work stigma and poorer mental health outcomes adds to previous research highlighting the role that internalized sex work stigma plays on HIV outcomes among cisgender women sex workers living with HIV in the DR (Carrasco et al., 2018; Zulliger et al., 2015). Though some interventions have focused on reducing internalized sex work stigma among cisgender female sex workers (Kerrigan, Barrington, Donastorg, Perez, & Galai, 2016; Kerrigan et al., 2019; Lippman et al., 2012), this has primarily occurred within the context of improving HIV outcomes. These multi-level interventions have included individual level counseling and health education, peer navigation, provider training, enhanced clinical services, and community mobilization and activities (Kerrigan et al., 2016; Kerrigan et al., 2019; Lippman et al., 2012). The current findings demonstrate that it is also important to focus on reducing internalized sex work stigma to promote mental health among trans women sex workers. Additional research is needed to explore how sex work stigma is internalized among trans women, and the mechanisms through which internalized sex work stigma influences their mental health.

Surprisingly, HIV stigma was not associated with depression or anxiety symptoms in any of the models, and trans stigma was only associated with increased anxiety symptoms in the unadjusted model. In the DR, there is a law (Ley de VIH/SIDA No. 135–11) that provides non-discrimination protections to people living with HIV. Protective policies are associated with improved mental health among trans people in other settings, such as the US (Du Bois, Yoder, Guy, Manser, & Ramos, 2018; Perez-Brumer, Hatzenbuehler, Oldenburg, & Bockting, 2015), and therefore, this policy may contribute to some of the differences in the associations between HIV stigma and mental health. However, it is important to note that this law is not always enforced, and that HIV stigma is still pervasive in the DR (Barrington et al., 2017).

Some differences in the associations across types of stigma may also result from differences in the visibility of identities. For example, HIV status may be less visible to others than gender identity, depending on an individual’s gender presentation. Trans women sex workers living with HIV in the DR have described relentless harassment and abuse due to their physical presentation as part of their daily lives (Barrington et al., 2020). The way in which someone is perceived by others (e.g., family, friends, community members) may influence how and if they experience enacted stigma, which could result in varied associations with mental health outcomes, including differences in internalized and anticipated stigma as well (Goffman, 1963).

Finally, differences in findings across types of stigma may also have resulted from differences in measures. For example, the measure of trans stigma captured multiple aspects of stigma in one scale, while the HIV and sex work stigma measures had separate scales measuring different aspects of stigma (i.e., anticipated, enacted, and internalized). It is possible that a more nuanced trans stigma scale (with multiple sub-scales) may be more appropriate for examining associations between trans stigma and mental health. However, while some trans stigma scales (e.g., the Gender Minority Stress Theory scales developed by Testa and colleagues (2015)) capture multiple aspects of trans stigma, these scales were developed within the context of the United States and may not apply to the experiences of trans women in the DR. The development of additional measures of trans stigma, based within the context of LMICs, is warranted.

Differences in results across types of stigma highlight the challenges of measuring and analyzing intersectional stigma. Although we measured multiple types of stigma as separate constructs, these types of stigma are co-occurring and mutually reinforce each other (Turan et al., 2019; Wesp, Malcoe, Elliott, & Poteat, 2019), and therefore measures may be simultaneously capturing multiple types of stigma. For example, in the DR, both trans stigma and HIV stigma result in challenges with finding employment beyond sex work (Barrington et al., 2017). This experience can potentially result in the assumption that someone is a sex worker if they also identify as trans. These overlaps can make it difficult for someone to discern if they are experiencing stigma because they are trans, because they are engaging in sex work, or because they are living with HIV. Future research should continue to explore the implications of assessing different components of intersectional stigma separately, or if intersectional stigma may be better measured as a scale that simultaneously captures multiple types of stigma (Turan et al., 2019).

Social cohesion was not particularly high or low. However, similar to previous research among trans women sex workers in the DR (Barrington et al., 2020; Milner et al., 2019), mistrust (an individual scale item) was high, with 77% of participants reporting that they did not trust the majority of other trans women sex workers who they know. We also found that trans women who experienced more social cohesion were less likely to experience depression or anxiety symptoms, which highlights the importance of social cohesion as a possible mechanism for resilience among trans women sex workers in the DR. In other countries, including in Latin America and the Caribbean, social cohesion and community connectedness have been identified as being associated with improved health for trans populations, and may serve as an important intervention mechanism for improving multiple health outcomes (especially HIV) among trans women (Austin & Goodman, 2017; Bockting et al., 2013; Matsuno & Israel, 2018; Perez-Brumer et al., 2017; Sherman, Clark, Robinson, Noorani, & Poteat, 2020). The current findings demonstrate that trans women sex workers in the DR may benefit from improved social cohesion, especially as a way to promote mental health; however, interventions need to consider the challenges of high mistrust within this community. The evaluation of a multi-level HIV intervention with trans women sex workers living with HIV in the DR found that individual-level components (e.g., counseling) had greater participation from trans women than interpersonal and community intervention components (e.g., peer navigation, community mobilization) due to low social cohesion and high mistrust (Barrington et al., 2020). Additional research is needed to understand how to overcome challenges with intervention participation resulting from low baseline levels of social cohesion.

While stigma and social cohesion variables had similar results with both mental health measures (depressive symptoms and anxiety symptoms), the illicit drug use covariate had distinct findings across the two outcomes. In general, substance use among this sample was high, with more than 50% of participants reporting illicit drug use in the past 6 months and more than 50% reporting binge drinking at least weekly in the past 6 months. Drug use was associated with having more anxiety, but was not significantly associated with depression. It is possible that drug use was used as a coping mechanism to manage anxiety (rather than depression). This may be especially true if participants were using drugs while engaging in sex work, since the potential violence that may be experienced during sex work may increase anxiety. Furthermore, since findings are cross-sectional, it is also possible that using drugs increased anxiety, especially if participants had concerns about the legal consequences of drug use. Additional research to better understand the associations between drug use and depression and anxiety is warranted.

Limitations.

There were some limitations with this study. Data are cross-sectional, so no causal effects can be determined. The lack of a randomly selected sample could also limit generalizability of results. The sample of 100 trans women was exceptionally large given the eligibility criteria of identifying as trans, engaging in sex work, and living with HIV; however, to increase the power to detect significance (Peduzzi et al., 1996), we limited the number of covariates included in the models. The small sample size also limited our ability to analyze the nuances of how HIV, sex work, and trans stigma intersect. For example, with a larger sample size, we may have been able to test a mediation or moderation analysis or conduct a latent class analysis to determine how these multiple types of stigma interact with each other.

Conclusions.

These findings demonstrate the importance of internalized sex work stigma on the mental health of trans women sex workers in the DR, and the need for future research and intervention to explore how to reduce internalized sex work stigma, as a mechanism for improving mental health. This study also highlights the important role of social cohesion as a social determinant of mental health; mental health promotion interventions in the DR may benefit from considering mechanisms for promoting social cohesion (e.g., community mobilization interventions and trans activism interventions). Additional research is needed to identify how to appropriately measure, analyze, and interpret intersecting forms of stigma.