Outness, Discrimination, and Psychological Distress Among LGBTQ+ People Living in the Southern United States

Abstract

Purpose:

Lesbian, gay, bisexual, transgender, queer, and other sexual and gender minority (LGBTQ+) people face mental health disparities. These disparities are amplified in the Southern regions of the United States. This study assessed the role of outness, discrimination, and other demographic variables on possible serious mental illness (SMI) among LGBTQ+ Southerners.

Methods:

This study used data from the 2017 LGBT Institute Southern Survey, a cross-sectional convenience sample of 6502 LGBTQ+ adults living in 14 Southern states. Multivariable logistic regression was performed to examine differences between those with and without possible SMI.

Results:

Outness was associated with a lower likelihood of possible SMI (odds ratio [OR]: 0.696, 95% confidence interval [CI]: 0.574–0.844, p = 0.001), especially when controlling for discrimination in the past 12 months (OR: 0.693, 95% CI: 0.576–0.834, p ≤ 0.001) and lifetime discrimination (OR: 0.678, 95% CI: 0.554–0.829, p = 0.001). Lifetime discrimination was associated with a higher likelihood of possible SMI (OR: 1.413, 95% CI: 1.034–1.932, p = 0.033), as was discrimination experienced in the past 12 months (OR: 1.626, 95% CI: 1.408–1.877, p ≤ 0.001). Black/African American respondents had the lowest percentage of possible SMI (21.0%) compared with other races, despite having lower or comparable rates of outness.

Conclusion:

These results indicate a possible promotive effect of outness against possible SMI among LGBTQ+ Southerners, as well as possible promotive group-level factors among Black/African American LGBTQ+ Southerners. Policies and interventions that address discrimination against LGBTQ+ Southerners should be expanded, and future research should address how the relationships between outness, discrimination, and mental health outcomes may vary by subgroup.

Introduction

An association exists between the unmet health needs of lesbian, gay, bisexual, transgender, queer, and other sexual and gender minority (LGBTQ+) people, their mental health, physical health, and morbidity.^1–5 The literature shows mixed results about if, when, how, and for whom outness is promotive versus a risk factor for mental health experiences and outcomes. Geographic region, local context, and exposure to discrimination affect how outness is related to mental health.

Existing evidence suggests that outness can have a promotive effect on the mental health of cisgender gay and lesbian people through a variety of mechanisms, such as increasing health service utilization,⁶ reducing substance use,⁷ facilitating positive identity formation,⁸ and increasing family and social support.^9,10 Conversely, outness often negatively impacts or does not affect these outcomes among bisexual and transgender people.^6,7,10 The promotive or negative impacts of outness vary by race^11–13 and contextual factors, such as rural versus urban location,¹⁴ and may not scale linearly with how “out” a particular person is.¹⁵

Although more than one third of LGBTQ+ people in the United States live in the South, very little research examines their experiences in this region of the country.¹⁶ As demonstrated by the Campaign for Southern Equality's 2019 Southern LGBTQ Health Survey, this is particularly concerning considering that LGBTQ+ people experience increased barriers to positive health outcomes and so do Southerners regardless of gender and sexuality. Therefore, LGBTQ+ Southerners are doubly burdened in this regard.¹⁷ The disparities were particularly pronounced for bisexual people, trans people, respondents between 18 and 24 years old, and respondents with lower incomes.

LGBTQ+ people experience a multitude of health disparities related to elevated levels of psychological distress.^18–22 Estimates of the percentage of sexual minority populations with a serious mental illness (SMI) range from 5.0% for lesbian women¹⁸ to 49.2% for men who have sex with men and use multiple drugs.²³ The mental health experiences of LGBTQ+ people vary considerably by sexuality,^18,24 gender,^4,5,18 age,^{5,13,15,25,26} national geography,^{3,16,24,27–31} and other factors. Although exact estimates of prevalence vary based on the symptoms or psychological definitions used in each study, the literature has robustly established mental health disparities among LGBTQ+ people. Research concerning bisexual and transgender populations and between different races is particularly lacking.^32–34

Minority stress theory suggests that experiences of discrimination and stigma become embodied through stress responses, which have lasting effects on health.^3,22,24 Stigma occurs at the individual, collective, and structural levels^35–37 and is a fundamental cause of health inequities.^38,39 Among LGBTQ+ people, stigma operates through a variety of interrelated social mechanisms, including perceived discrimination²⁶ and victimization due to identity.^33,40

Findings are mixed about the relationship between discrimination, outness, and mental health. Most research indicates that outness is related to better mental health outcomes.^{4,13,15,26,33,41} At the same time, people who are out are more likely to experience discrimination due to the visibility of their identities.^28,42 For instance, one study found that people who were more out at work experienced less indirect heterosexism (such as being assumed to be heterosexual), but more direct heterosexism (such as anti-gay jokes).³⁴ Thus, outness' promotive effect is contingent on local context.

Discrimination contributes more to the mental health disparities among LGBTQ+ populations in locations that do not have anti-discrimination laws or policies.^{27,28,30,41,43} Despite majority support for anti-discrimination policies in the Southern United States, statewide laws are not in place in most Southern states.⁴⁴ Consequently, the South continues to have lower levels of social acceptance of LGBTQ+ people than other regions of the country.^{27–30,41,43} The Movement Advancement Project tracks LGBT laws and policies across the United States.⁴⁴ Most Southern states continue to score negative policy tallies, which suggests that the social environments for LGBTQ+ people in the South are more hostile.^44,45 Overall, place matters when it comes to LGBTQ+ people's lives and experiences, particularly related to health.^9,37,38,41

This study adds to the limited research on LGBTQ+ people and health in the Southern United States and expands research to better understand the relationship between outness, discrimination, and psychological distress. We extend knowledge on LGBTQ+ Southerners' outness and mental health by analyzing a large sample (n = 6502) of survey respondents to explore the relationships between LGBTQ+ Southerners' ability to be open about their gender and sexual identities with close family and friends and the possibility of having an SMI. We pay particular attention to race, gender variance, and bisexuality, as these identities have been shown to impact the relationship between outness and health outcomes. This study is unique in that it has a sample size larger than many other LGBTQ+ Southerner-specific surveys,^17,46 and it provides specific analyses of Southern experiences for LGBTQ+ people.⁴⁷

Methods

The 2017 LGBT Institute Southern Survey is a cross-sectional sample of LGBTQ+-identified adults living in 14 Southern states: Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia. The goal of the original survey was to amplify voices of and highlight issues affecting LGBTQ Southerners. The survey received approval from the Georgia State University Institutional Review Board (IRB) and was administered online and resulted in a sample of 6502 participants, each of whom provided informed consent. Participants were recruited through social media, online networks, and Facebook advertisements. This secondary study received an exemption from the Georgia State University IRB because it uses previously collected, de-identified data.

Study variables

Demographic variables

Participant demographic variables collected were age, gender, sexual orientation, race/ethnicity, income, and educational attainment. Sexual identity was measured by the question, “Would you say you are …”: “Heterosexual or straight,” “Lesbian,” “Gay,” “Bisexual,” and “Some other sexual identity, please specify.” Participants were asked their sex assigned at birth, if they are a transgender person, and how they think of their gender now (with answer options of man, woman, genderqueer, and other). For this article, gender was measured with four mutually exclusive categories: “Cis man,” “Cis woman,” “Transgender,” and “Other.”

Age was categorized into 10-year groupings. Race/ethnicity was based on self-report: “Non-Hispanic White,” “Black/African American,” “Hispanic/Latino/Latina/Latinx,” and “Other.” In this article, we use the term Hispanic/Latinx while acknowledging that others prefer other terms such as Latino/Latina/Latine. Household income was categorized into five quintiles. Educational attainment was categorized into four groups: “High school, GED, or less”; “Some college or 2-year degree”; “4-year degree”; and “Graduate/professional/doctoral degree.”

Psychological distress

Psychological distress was measured using the Kessler 6 Psychological Distress Scale (K6).⁴⁸ The K6 scale asks participants how often they felt six different difficult feelings (hopeless; nervous; restless or fidgety; so depressed that nothing could cheer you up; that everything was an effort; worthless) during the past 30 days, graded on a 5-point Likert scale ranging from “All of the time” to “None of the time.” Based on clinical standards, a sum of the six questions was created, resulting in a variable with a range of scores from 0 to 24, with a score of 13 or higher considered high risk for possible SMI.

Discrimination

Discrimination that participants experienced due to LGBTQ+ identity was measured with nine discrimination questions (Table 1). Based on the nine questions, two dichotomous variables were created for either having experienced discrimination or not in the past 12 months and over respondents' lifetimes.

Table 1.

Discrimination Due To Sexual and Gender Minority Identity Questions Used for Discrimination Scale

	Percent lifetime	Percentage in the past 12 months	n
1. Been threatened or physically attacked	40.5	6.1	4210
2. Been subject to slurs or jokes	79.6	28.9	4208
3. Received poor service in restaurants, hotels, or other places of business	44.1	14.9	4207
4. Been made to feel unwelcome at a place of worship or religious organization	57.4	15.3	4205
5. Been treated unfairly by an employer in hiring, pay, or promotion	27.7	6.4	4201
6. Been rejected by a friend or family member	66.6	19.9	4199
7. Been unfairly stopped, searched, questioned, physically threatened, or abused by the police	10.6	2.0	4204
8. Been prevented from moving into a neighborhood because the landlord or realtor refused to sell or rent to you a house or apartment	5.9	1.0	4207
9. Been denied care or treated unfairly by a health care provider	17.9	5.0	4207
Discrimination scales	90.5	45.7	4170

Outness

Outness was measured by the questions “How many people in each group below know you are [their self-identified sexual orientation or gender identity].” Groups included were “Immediate family you grew up with (mother, father, sisters, brothers, etc.),” “LGBT friends,” and “Straight, non-LGBT friends.” Possible answers included: “I have no people like this in my life,” “All know that I am,” “Most know that I am,” “Some know that I am,” “None know that I am.” An outness among close family and friends scale was created by summing the answers for each of the three groups. A dichotomous variable was created that distinguished between people who were out to all their family and friends and those who had family and friends who did not know they identify as a sexual or gender minority.

Of note, the outness scale was also dichotomized into upper and lower halves, and the same analyses were performed with similar results. Additional correlation analyses between outness, discrimination, and psychological distress as continuous variables were calculated and yielded similar results. As such, these results are not presented in detail in this article.

Statistical analyses

SPSS statistical software was used to perform chi-square tests and multivariate logistic regression to evaluate possible SMI across different demographic and discrimination variables. Three regression models were performed: one controlling for lifetime discrimination and demographic variables, another controlling for discrimination experienced in the past 12 months and demographic variables, and a third controlling only for demographic variables.

Results

The majority of respondents were younger, with 37% between ages 18 and 29 years. Forty-three percent of the sample identified as cis women, 35% as cis men, 17% as transgender, and 5% as some other gender. There was significant diversity of sexual identities, with 33% identifying as gay men, 24% as lesbian, 21% as bisexual, and 15% as another sexuality. Respondents were mostly non-Hispanic White (74%), with 7% of the sample identifying as Black/African American and 6% identifying as Hispanic/Latinx. The remaining 6% identified as other races or ethnicities. The sample was relatively wealthy and well educated, with 17% of the population in the highest fifth quintile of household income and 36% of respondents having a graduate/professional degree.

Only 0.74% of the sample (30 individuals) reported being out to no one, whereas 39.1% reported being out to all their family and close friends. Most participants (90.5%) had experienced discrimination of some kind in their lifetime due to LGBTQ+ identity, and 45.7% had experienced discrimination in the past 12 months due to LGBTQ+ identity. Approximately a quarter of respondents (23.8%) had a possible SMI.

Table 2 presents bivariate analyses of selected variables by possible SMI status. Transgender respondents and those who identified as some other gender had higher percentages of possible SMI (40.3% and 39.3%, respectively) compared with cisgender men (16.3%) and cisgender women (21.8%). A higher percentage of bisexual people (34.8%) had possible SMI compared with lesbian (17.3%) and gay (15.1%) people, although “other” sexualities had the highest percentage (37.6%).

Table 2.

Selected Variables by Possible Serious Mental Illness Status

Predictor	No SMIa		SMIa
	n	%	n	%
Age*, years
18–29	988	62.3	598	37.7
30–39	816	76.5	251	23.5
40–49	623	83.1	127	16.9
50–59	599	89.3	72	10.7
60–69	368	90.0	41	10.0
70 or older	109	93.2	8	6.8
Gender*
Man	1342	83.7	261	16.3
Woman	1561	78.2	435	21.8
Trans	475	59.7	320	40.3
Other	125	60.7	81	39.3
Sexual orientation*
Heterosexual	208	84.9	37	15.1
Lesbian	964	82.7	202	17.3
Gay	1270	82.7	262	17.1
Bisexual	617	65.2	329	34.8
Other	444	62.4	267	37.6
Race/ethnicity*
Non-Hispanic White	2888	77	865	23.0
Black/African American	229	79.0	61	21.0
Hispanic	181	69.3	80	30.7
Other	201	68.8	91	31.2
Income*
Low	430	61.4	270	38.6
Second	317	74.4	109	25.6
Third	463	78.7	125	21.3
Fourth	445	84.1	84	15.9
Fifth	466	89.4	55	10.6
Educational attainment*
High school, GED or less	179	61.3	113	38.7
Some college/2-year	843	67.8	401	32.2
4-Year degree	1052	75.8	336	24.2
Grad/prof/doctoral	1426	85.3	246	14.7
Lifetime discrimination*
Yes	2803	75.0	934	25.0
No	318	82.4	68	17.6
Discrimination—12 months*
Yes	1265	66.9	627	33.1
No	1856	83.2	375	16.8
Out to family and friends*
All know	1412	83.6	276	16.4
Not all know	1844	70.4	776	29.6

^* Significance <0.001 Pearson's chi-square.

^a SMI measured with the K6 Psychological Distress Scale.

GED, General Education Development; SMI, serious mental illness.

Races other than Black/African American and White, including Asian and Native American people, had the highest percentage of people with possible SMI (31.2%). Black/African American people had the lowest percentage of possible SMI (21.0%), just narrowly less than non-Hispanic White people (23.0%). People who had experienced discrimination in the past 12 months had higher percentages of possible SMI (33.1% vs. 16.8%, respectively). Similar results were seen among those who experienced discrimination in their lifetime (25.0% vs. 17.6%, respectively). A larger percentage of those who were not out to all their family and friends had a possible SMI (29.6%) compared with those who were out to all their family and friends (16.4%).

Logistic regression analysis (Table 3) showed that those who were out to all their family and friends had significantly lower likelihood (odds ratio: 0.696, 95% confidence interval [CI]: 0.574–0.844) of having an SMI compared with those who were not fully out when controlling for demographic variables. When controlling for demographic variables and lifetime discrimination, those who were out to all their family and friends had 0.678 times the odds of having an SMI compared with those who were not out to all their family and friends (95% CI: 0.554–0.829). When controlling for discrimination in the past 12 months and demographic variables, those who were out to all their family and friends had 0.693 times the odds (95% CI: 0.576–0.834) of having an SMI compared with those who had restricted outness.

Table 3.

Regression Analysis for Effects of Outness on the Likelihood of Possible Serious Mental Illness

	Model 1		Model 2		Model 3
	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
Age	0.737 (0.698–0.779)	<0.001	0.737 (0.698–0.779)	<0.001	0.750 (0.710–0.792)	<0.001
Household income	0.820 (0.750–0.897)	<0.001	0.822 (0.752–0.898)	<0.001	0.825 (0.758–0.898)	<0.001
Educational attainment	0.799 (0.743–0.859)	<0.001	0.795 (0.740–0.855)	<0.001	0.807 (0.748–0.870)	<0.001
Hispanic identity	1.186 (0.905–1.554)	0.213	1.200 (0.914–1.574)	0.186	1.193 (0.912–1.562)	0.196
Other race identity	1.245 (0.929–1.669)	0.137	1.263 (0.946–1.687)	0.110	1.240 (0.923–1.665)	0.146
Trans or other gender identity	1.589 (1.368–1.845)	<0.001	1.566 (1.346–1.822)	<0.001	1.493 (1.277–1.744)	<0.001
Other sexual orientation identity	1.486 (1.242–1.778)	<0.001	1.509 (1.261–1.805)	<0.001	1.514 (1.262–1.815)	<0.001
Bisexual identity	1.569 (1.332–1.848)	<0.001	1.597 (1.359–1.878)	<0.001	1.622 (1.374–1.914)	<0.001
Outness	0.696 (0.574–0.844)	0.001	0.678 (0.554–0.829)	0.001	0.693 (0.576–0.834)	<0.001
Lifetime discrimination	—		1.413 (1.034–1.932)	0.033	—
Discrimination in the past 12 months	—		—		1.626 (1.408–1.877)	<0.001
−2 Log likelihood	2518.706		2478.838		2459.351
Cox and Snell R Square	0.136		0.139		0.146
Nagelkerke R Square	0.204		0.209		0.219

CI, confidence interval; OR, odds ratio.

When controlling for demographic characteristics and outness, those who experienced discrimination in the past 12 months had 1.626 times the odds of having an SMI compared with those who did not experience discrimination in the past 12 months (95% CI: 1.408–1.877). When controlling for demographic characteristics and outness, those who experienced discrimination in their lifetime had 1.413 times the odds of having an SMI compared with those who did not experience discrimination in the past 12 months (95% CI: 1.034–1.932).

Of note, all analyses were also completed using the secondary outness scale that measured higher versus lower outness among family and friends. Higher outness was associated with better mental health, indicating being out to most (as opposed to all) family and close friends may similarly be a promotive factor for mental health.

Discussion

This study provides insight into some of the factors affecting the mental health disparities that LGBTQ+ people face. Namely, the association between outness and possible SMI suggests that social scenarios that do not allow for outness are negatively affecting LGBTQ+ people's mental health. This article explores these factors in LGBTQ+ Southerners, who have a higher prevalence of possible SMI compared with national estimates.⁴⁹ Consistent with the existing literature on LGBTQ+ Southerners,^17,46 bisexual people, transgender people, and people who identified their sexuality or gender as “other” had higher percentages of possible SMI.

Although bisexual people experienced less discrimination than lesbian and gay respondents, bisexual respondents had worse mental health outcomes. This is likely related to the fact that only 15% of bisexual people were out to all their family and friends, compared with 51.4% of lesbian women, 53.4% of gay men, and 21.7% of people with other sexualities. Thus, bisexual people were less likely to have social support related to their sexual identity. Being out to friends and close family is a promotive factor against SMI. Outness promotes better mental health through social support, while also putting one at risk for discrimination, which can limit the promotive effect.

Transgender respondents had similar percentages of being out to all their family and friends (17.7%) as bisexual people and also had worse mental health outcomes compared with cisgender gay and lesbian respondents. Transgender people were also more likely to experience discrimination compared with their cisgender lesbian, gay, and bisexual counterparts. Researchers and clinicians developing interventions with LGBTQ+ Southerners should remember that the relationship between outness and improved mental health outcomes is contextually dependent, especially because discrimination may be experienced differently in varying contexts.

Black/African American LGBTQ+ Southerners experienced less psychological distress compared with other racial/ethnic groups despite having slightly lower outness. A total of 33.6% of Black/African America respondents were out to all their family and friends compared with 40.4% of non-Hispanic White respondents, 36.5% of Hispanic/Latinx respondents, and 30.1% of other races. It is possible that being Black/African American is a promotive factor for LGBTQ+ Southerners because of a larger proportion of Black/African American people in the South, compared with other parts of the United States, and this could foster greater social support via racial solidarity within the Southern Black community.

Identified resilience factors among Black LGBTQ+ people include social support through religion, spirituality, and integration of intersecting marginalized identities to resist and persist in the face of oppression.⁵⁰ It is possible that these factors may help protect against psychological distress even when Black Southerners are not out about their LGBTQ+ identity—for example, in our sample, only 33.6% of Black/African American respondents were fully out but demonstrated better outcomes.

Further research should investigate community- and individual-level factors among Black/African American LGBTQ+ Southerners, such as community cohesion and resistance to internalized stigma. Incorporating perspectives and methods on mental health resilience could help understand how outness may impact mental health differently across different racial or ethnic groups across the Southern United States.^51–53

Limitations

There were several limitations of the study. The sample is a convenience sample and may not be representative of the general LGBTQ+ population in the South, as evidenced by a low proportion (6.6%) of Black/African American respondents. The online survey format provides convenience to participants but could exclude individuals who do not have a computer or smartphone with internet access. This bias could be one reason the sample has a higher income and education than the general population, which may also cause an underestimation of the prevalence of SMI. In addition, all the measurements were self-reported, which could produce recall bias.

It is possible that because the K6 measurement scale was designed for a nationally representative sample, the measurements on specific subpopulations such as LGBTQ+ individuals or individuals of different races are inaccurate.⁴² In addition, simplifying mental health challenges into an SMI scale loses nuances and variety of mental health information. Finally, because the data were cross-sectional, causality cannot be assumed.

Future directions

There is a need for more population-based studies of LGBTQ+ individuals to be able to generalize these findings⁵⁴ and compare different regions of the United States. In addition, research should strive to better understand the connections between outness, psychological distress, and psychological well-being.¹⁶ Factors that affect the relationship between outness and mental health, such as social support and self-acceptance across race- and gender-based communities should be explored further to create interventions. These interventions should be planned and led by members of the affected communities and capitalize on the benefits of outness. For instance, collective action has been found to buffer the relationship between discrimination and internalized heterosexism,³³ and community connectedness is associated with decreased anxiety among Southerners.²⁶

Anti-discrimination policies should be universal due to discrimination mitigating some of the health benefits of being out.³⁴ Acknowledgement of and education about LGBTQ+ people in schools and workplaces could help people feel safe enough to come out, as well as reduce discrimination.^34,43 Sexuality and gender protections could reduce bullying and discrimination and subsequently improve mental health,⁵⁵ although punitive laws often fail to prevent harm and disproportionately hurt marginalized people.⁵⁶ Clinical interventions should consider the benefits of identity formation and coming out.⁵³ Factors that affect the relationship between outness and mental health, such as social support and strong identity, should be explored further to create interventions that capitalize on the benefits of outness.

Conclusion

This study provides a critical look into mental health disparities faced by LGBTQ+ Southerners and the effect that discrimination and outness may have on mental health outcomes. There was a negative association between outness and possible SMI and a positive association between discrimination and possible SMI. Interventions to reduce discrimination and allow people to disclose their sexual and gender identities could reduce mental health disparities.

Authors' Contributions

J.A.C. co-conceptualized the article, developed methodology, performed the statistical data analysis, drafted the original article, and revised it critically along with other co-authors. A.B. assisted with statistical methods and analyses, interpretation of data, and writing and revision of the article. B.A.R. assisted with revising the article critically along with other co-authors. R.R. helped with data curation. E.W. co-conceptualized the article, assisted with statistical methods and analysis and interpretations of data, and helped with critically revising the article. Each author reviewed and approved the final version of the article for publication and is accountable for the accuracy and integrity of the research.

Disclaimer

Primary data collection was performed by the LGBTQ Institute in collaboration with Georgia State University. The data set was obtained from Dr. Eric Wright.

Author Disclosure Statement

A.B. works part-time as a clinical trials scientific advisor with IQVIA. No competing financial interests exist.

Funding Information

This article was not funded. However, over the course of this study, A.B. was funded by a T-32 training grant in Gender, Sexuality, and Health provided by the National Institute of Child Health and Human Development (NICHD) at the National Institutes of Health (NIH) (Grant No. 5T32Hd049339-15). He was also funded by the Graduate Research Fellowship Program (GRFP) from the National Science Foundation (NSF) (Grant No. DGE 2036197).