Associations between Living Arrangement and Sexual and Gender Minority Stressors among University Students Since the Start of the COVID-19 Pandemic

Abstract

Objective:

To examine the associations between COVID-19-related living arrangements and sexual and gender minority (SGM)-related stressors (i.e., identity concealment and familial rejection).

Participants:

N= 478 SGM university students (M_age=22 years, SD=4.00)

Methods:

SGM university students were surveyed cross-sectionally between May-August 2020 regarding SGM-related stressors and living arrangements since the start of COVID-19.

Results:

Approximately half (48.7%) of the sample reported a living rearrangement to their parents’ home due to COVID-19. Living rearrangement to parents’ homes was associated with an increased degree of identity concealment (β[95%C.I.]=0.62[0.10, 1.15]; p=0.020) and familial rejection (β[95%C.I.]=1.56[0.72, 2.41]; p<0.001) since the start of COVID-19 compared to stably living without parents (34.3%). Stably living with parents (17.0%) was not associated with increased degree of SGM-related stressors compared to experiencing a living rearrangement.

Conclusions:

Stakeholders must consider the unique identity-related vulnerabilities of SGM students living with parents and who experience living rearrangements due to COVID-19.

Introduction

In response to the Coronavirus Disease 2019 (COVID-19) pandemic, United States (U.S.) universities ceased in-person instruction and physically closed campuses in 2020, including campus housing, causing some students to experience a living rearrangement by having to go back to living with their parents. Unfortunately, such closures and virtual operations likely had collateral consequences on the mental health of university students as a result of reduced access to community, social, and mental health support.^1–4 Sexual and gender minority (SGM) university students specifically face unique risks for experiencing identity-related stressors (e.g., SGM identity-related identity concealment and familial rejection) and increased mental health burden if they transitioned to homes with unaffirming parents.^2,5 Given the significant pandemic-related vulnerabilities faced by SGM^2,5,6 and university students,^3,4,7 it is important to examine associations between identity-related stressors and living rearrangements among SGM university students since the start of COVID-19.

Universities sometimes provide SGM students independence from stigmatizing home environments and the ability to live openly and authentically, often for the first time.⁸ Many institutions also have supportive and identity-affirming services on campus that aim to improve the overall success and wellbeing of SGM students, such as LGBTQ student centers, offices of diversity and equity, and mental health services. For some, living rearrangement from universities to parents’ homes as a result of COVID-19 could position SGM students in unaffirming and unsafe home environments that may threaten their mental health and wellbeing by forcing them to conceal their identities, and subject them to familial rejection as a result of their SGM identities.^2,5,9 Similarly, some SGM students who are already living with unaffirming parents and are forced to spend more time with them as a result of COVID-19 are at risk for heightened mental health burden.¹⁰ However, the mechanisms explaining such increases in mental health risk have not yet been investigated.

Limited research has examined the secondary mental health impacts of the COVID-19 pandemic on SGM university students in the U.S. Minority Stress Theory¹¹ posits that proximal (e.g., identity concealment) and distal (e.g., familial rejection) SGM-related stressors significantly impact the mental health of SGM persons. Using this framework, our study aimed to examine whether COVID-19-related living arrangements among SGM university students were associated with different degrees of SGM-related stressors (i.e., SGM-related identity concealment and familial rejection). We hypothesized that SGM university students who experienced a living rearrangement to their parents’ homes would experience a greater degree of SGM-related stressors since the start of COVID-19 compared to those stably living with or without their parents. Findings from our study could inform research, practice, and policies to reduce mental health inequities among SGM university students amid COVID-19 and beyond.

Materials and Methods

Between May 27^th and August 14^th, 2020, a non-probability cross-sectional survey was conducted online to explore the mental health impacts of COVID-19 on SGM university students. Participants were recruited using an electronic recruitment flyer via social media and email campaigns at universities across the U.S. The recruitment flyer read, “Are you an LGBTQ+ college student? Is COVID-19 impacting you? We need to hear from you!”. The flyer was distributed via multiple social media platforms (i.e. Facebook, LinkedIn, and Twitter). We also recruited through email campaigns internally at the University of Maryland (e.g. student organizations, professors, peers, administration, leadership, campus centers), within co-author internal and external professional networks (peers and colleagues), at historically Black colleges and universities and Hispanic serving institutions (offices of diversity, student affairs, or student services), and lesbian, gay, bisexual, transgender, and queer or questioning (LGBTQ) student offices and centers across the U.S.

Eligibility criteria included being 1) 18 years or older, 2) a full-time undergraduate or graduate student at a U.S. institution, and 3) identifying as a SGM person. Before completing the survey, participants provided electronic informed consent. Survey duration was approximately 20–25 minutes and participants were offered the chance to enter into a drawing for a $50 Amazon gift card upon completion. Institutional review board approval from the University of Maryland was obtained prior to commencing data collection.

To protect data integrity against robots or careless people entries, we implemented three ex-ante attention-check items in separate places on the survey.¹² Specifically, we implemented one “explicitly instructed response” item, which instructed participants to select a specific answer option (requires comprehension and response), and two “bogus” items, which are items added on to existing scales to test for participant attention (requires comprehension, retrieval, judgement, and response).¹² Failing to correctly answer at least two of the three items was established as the threshold indicative of a potential robot or careless person entry. Of the entire study sample, only two participants failed to pass the established threshold (99.7% pass rate) and were thus removed from analysis.

Living arrangements

Participants were asked about their living arrangements prior to and during COVID-19. SGM students were classified into three groups: stable arrangement living without parents before and during COVID-19 (referent group), stable arrangement living with parents before and during COVID-19, and rearrangement (not living with parents before COVID-19, but presently living with parents during COVID-19). Students that did not fit these living arrangements were removed from the sample (n=9).

SGM-related stressors

We ascertained degree of SGM-related identity concealment and familial rejection since the start of COVID-19. SGM-related identity concealment was measured using a cumulative score of 7-items adapted from the Outland¹³ and Balsam et al¹⁴ scales, (range=0–7, α=.843) asking, “Have you felt this way more often since the start of the COVID-19 pandemic? Yes (Score=1) or No (Score=0)” for each individual item (e.g. Avoided telling people about certain things in my life that might imply I am LGBTQ; Watched what I said and did around heterosexual people). Higher cumulative scores indicated a greater degree of identity concealment since the start of COVID-19.

SGM-related familial rejection was measured using a cumulative score of 17-items adapted from the Schrager et al¹⁵ and Balsam et al¹⁴ scales (range=0–17, α=.899) asking, “Has this happened to you more often since the start of the COVID-19 pandemic? Yes (Score=1) or No (Score=0)” for each individual item (e.g. Being rejected by my mother for being LGBTQ). Higher cumulative scores indicated a greater degree of familial rejection since the start of COVID-19.

Model Covariates

Previous literature informed our selection of model covariates age,¹⁶ sex assigned at birth,¹⁷ gender identity,^18,19 sexual identity,^17,20 ethnicity,²⁰ race,^19,21 nativity,²² parental financial dependence,^23,24 and familial outness²⁵ as these have been associated with SGM-related stressors and SGM stress-related health outcomes in existing epidemiological and SGM health research. Gender identity and sex assigned at birth were collected using the gold standard 2-step question process,²⁶ where participants are asked their gender identity and sex assigned at birth using separate questions. Parental financial dependence was measured with the item, “How financially dependent are you on a primary caregiver?” with Likert responses ranging from “Not at all financially dependent (value=1)” to “Entirely financially dependent (value=5)”. Outness and identity concealment are similar but distinct constructs, where outness describes who knows about one’s SGM identification, and identity concealment describes a process of avoiding social situations that engage in topics related to SGM identification because of anticipated stigma. Average familial outness was measured by asking participants, “How many people in each group below currently know you are LGBTQ? 1) Parents, 2) Guardians, 3) Siblings, 4) Extended family members” with Likert responses ranging from “None know I am LGBTQ (value=1)” to “All know I am LGBTQ (value=4)” and the additional non-applicable option of “I currently don’t have people like this in my life.” The total score of familial outness among groups applicable to the participant, divided by the number of applicable groups determined average familial outness. For a full listing of covariate categories, see Table 1.

Table 1.

Demographic characteristics of N=478 sexual and gender minority-identifying university students stratified by living arrangement status during the Coronavirus Disease 2019 pandemic

	1. Stable Arrangement (without parents) n=164	2. Stable Arrangement (with parents) n=81	3. Rearrangement (no parents to parents) n=233			Tukey-Kramer Post-Hoc Analyses/ Stratified χ2 Analysesb
	n(%)	n(%)	n(%)	χ2 / Fa	p	p (1 vs 2)	p (1 vs 3)	p (2 vs 3)
Age (M±SD)c	25.0±5.1	21.0±2.5	20.2±1.6	101.9	<0.001	<0.001	<0.001	0.248
Sex assigned at birth				1.8	0.406	---	---	---
Male	38 (23.5)	20 (24.7)	44 (19.0)
Female	124 (76.5)	61 (75.3)	189 (81.1)
Gender				10.7	0.005	0.832	0.003	0.017
Cisgender	105 (64.0)	52 (64.2)	181 (77.7)
Non-cisgender	59 (36.0)	29 (35.8)	52 (22.3)
Race				1.8	0.781	---	---	---
White	114 (70.8)	52 (64.2)	162 (69.5)
Black	11 (6.8)	9 (11.1)	20 (8.6)
Something else	36 (22.4)	20 (24.7)	51 (21.9)
Ethnicity				7.1	0.029	0.028	0.875	0.012
Hispanic	22 (13.4)	20 (24.7)	30 (12.9)
Non-Hispanic	142 (86.6)	61 (75.3)	203 (87.1)
Nativity				1.7	0.427	---	---	---
U.S.-born	151 (92.1)	77 (95.1)	209 (90.5)
Foreign-born	13 (7.9)	4 (4.9)	22 (9.5)
Sexuality				7.5	0.114	---	---	---
Gay/Lesbian	57 (34.8)	27 (33.3)	71 (30.5)
Bisexual	41 (25.0)	25 (30.9)	88 (37.8)
Non-LGBd	66 (40.2)	29 (35.8)	74 (31.8)
Financial Dependence (M±SD)c	2.2±1.3	4.0±0.9	4.0±1.0	143.3	<0.001	<0.001	<0.001	0.879
Familial Outness (M±SD)c	2.7±1.1	2.5±1.1	2.4±1.0	3.9	0.020	0.304	0.015	0.792

^aχ² was calculated for categorical variables and F-values were calculated for continuous variables

^bTukey-Kramer Post-Hoc Analyses were conducted for significant ANOVA results from continuous variables and Stratified χ² Analyses were conducted for significant overall χ² results from categorical variables

^cMean ± Standard Deviation

^dNon-Lesbian, Gay, or Bisexual identified

Statistical analyses

Statistical analyses were conducted in SAS Studios^®. Frequencies, means, and bivariate associations of sample demographics were calculated by living arrangement status (Table 1). To answer our hypotheses, analysis of variance (ANOVA) were conducted to examine bivariate associations between living arrangements and SGM-related identity concealment and familial rejection (Table 2). Significant results were further examined utilizing Tukey-Kramer post-hoc analyses (Table 2). Next, we conducted unadjusted bivariate linear regression models assessing the impact of living arrangement (referent= ‘stable Arrangement (without parents)’), age (continuous), sex assigned at birth (referent= ‘male’), gender identity (referent= ‘cisgender’), race (referent= ‘white’ ), ethnicity (referent= ‘non-Hispanic’), nativity (referent= ‘U.S.’ ), sexual identity (referent= ‘gay/lesbian’), parental financial dependence (continuous), and familial outness (continuous) on outcomes identity concealment and familial rejection (Table 3). Variables reaching a significance level of p<0.05 in unadjusted bivariate regression analyses were then carried forward to adjusted multivariate linear regression models with identity concealment and familial rejection (Table 3). By controlling for potential confounding variables, we are able to provide estimates that are closer to the ‘true’ impact of living arrangement on degree of SGM identity concealment and family rejection.

Table 2.

Bivariate associations between sexual and gender minority-related stressor frequency since the Coronavirus Disease 2019 pandemic and living arrangement status among N=478 sexual and gender minority university students

	1. Stable Arrangement (without parents)	2. Stable Arrangement (with parents)	3. Rearrangement (no parents to parents)				Tukey-Kramer Post-Hoc Analysisb
	M±SDc	M±SDc	M±SDc	r 2a	F-Value	p-value	1 vs 2	1 vs 3	2 vs 3
Identity Concealment	1.3±1.9	2.1±2.3	2.2±2.2	0.042	10.21	P<0.001	P=0.012	P<0.001	P=0.873
Familial Rejection	1.5±2.7	2.9±3.9	3.4±3.0	0.056	13.86	P<0.001	P=0.012	P<0.001	P=0.520

^asemipartial eta-squared value for individual SGM-stressors

^bTukey-Kramer Post-Hoc Analyses were conducted for significant ANOVA results from continuous variables

^cMean ± Standard Deviation

Table 3.

Multivariate associations between sexual and gender minority-related stressor frequency since the Coronavirus Disease 2019 pandemic and living arrangement status among N=478 sexual and gender minority university students

	Identity Concealment					Familial Rejection
	β (95% C.I.)	p	adj. β (95% C.I.)	p	r 2a	β (95% C.I.)	p	adj. β (95% C.I.)	p	r 2a
Living Arrangement Status					0.042					0.054
Stable Arrangement (without parents)	Ref.	Ref.	Ref.	Ref.		Ref.	Ref.	Ref.	Ref.
Stable Arrangement (with parents)	0.77 (0.21, 1.33)	0.008*	0.52 (−0.11, 1.15)	0.104		1.35 (0.45, 2.27)	0.004*	1.37 (0.36, 2.39)	0.008*
Rearrangement (no parents to parents)	0.97 (0.54, 1.40)	<0.001**	0.62 (0.10, 1.15)	0.020*		1.80 (1.13, 2.48)	<0.001**	1.62 (0.77, 2.47)	<0.001**
Age					0.009					0.012
Years	−0.10 (−0.15, −0.05)	<0.001	−0.04 (−0.10, 0.02)	0.171		−0.19 (−0.27, −0.12)	<0.001**	−0.14 (−0.23, −0.04)	0.006*
Sex assigned at birth					---					---
Male	Ref.	Ref.	---	---		Ref.	Ref.	---	---
Female	0.17 (−0.31, 0.65)	0.483	---	---		0.76 (−0.00, 1.52)	0.050	---	---
Gender					---					---
Cisgender	Ref.	Ref.	---	---		Ref.	Ref.	---	---
Non-Cisgender	0.08 (−0.35, 0.51)	0.727	---	---		0.24 (−0.45, 0.93)	0.498	---	---
Race					---					---
White	Ref.	Ref.	---	---		Ref.	Ref.	---	---
Black	0.32 (−0.39, 1.04)	0.372	---	---		0.92 (−0.23, 2.06)	0.118	---	---
Something else	0.09 (−0.38, 0.56)	0.709	---	---		0.21 (−0.55, 0.96)	0.594	---	---
Ethnicity					---					---
Non-Hispanic	Ref.	Ref.	---	---		Ref.	Ref.	---	---
Hispanic	0.37 (−0.18, 0.91)	0.189	---	---		0.79 (−0.08, 1.66)	0.074	---	---
Nativity					0.010					0.015
U.S. Born	Ref.	Ref.	Ref.	Ref.		Ref.	Ref.	Ref.	Ref.
Foreign Born	0.86 (0.16, 1.57)	0.016*	0.44 (−0.22, 1.11)	0.192		1.65 (0.53, 2.78)	0.004*	1.19 (0.12, 2.26)	0.030*
Sexuality					---					---
Gay/Lesbian	Ref.	Ref.	---	---		Ref.	Ref.	---	---
Bisexual	−0.02 (−0.51, 0.46)	0.932	---	---		0.32 (−0.46, 1.09)	0.425	---	---
Non-LGBb	−0.03 (−0.50, 0.45)	0.905	---	---		0.59 (−0.16, 1.35)	0.123	---	---
Financial Dependence					0.000					0.012
Per unit increase	0.24 (0.07, 0.38)	0.001*	0.02 (−0.20, 0.17)	0.867		0.25 (0.02, 0.48)	0.032*	−0.39 (−0.68, -0.10)	0.009*
Average Familial Outness					0.092					0.040
Per unit increase	−0.73 (−0.91, −0.55)	<0.001**	−0.66 (−0.84, −0.48)	<0.001**		−0.87 (−1.17,−0.57)	<0.001**	−0.72 (−1.01, −0.42)	<0.001**

^asemipartial eta-squared value for individual SGM-stressors

^bNon-Lesbian, Gay, or Bisexual identified

^*p<0.05

^**p<0.001

Results

Sample demographic characteristics are reported in Table 1. Since the onset of COVID-19, our sample indicated an average increase of 1.9 (standard deviation (SD)=2.2) SGM-related identity concealment experiences and 2.6 (SD=3.5) familial rejection experiences. In bivariate tests (Table 2), living arrangement was found to be significantly associated with SGM-related identity concealment (p<0.001) and familial rejection (p<0.001) since the start of COVID-19, where those who experienced a living rearrangement due to COVID-19 or were stably living with their parents experienced a higher degree of SGM-related stress than those stably not living with their parents. In Tukey-Kramer post-hoc analyses, we found that those stably living with parents and those who experienced a living rearrangement did not statistically differ on degrees of SGM-related stressors since the start of COVID-19.

In unadjusted and adjusted regression models (Table 3; only adjusted reported in-text), those who experienced a living rearrangement due to COVID-19 experienced a significantly greater degree of SGM-related identity concealment since the start of COVID-19 compared to those stably living without their parents (β[95% C.I. (95% Confidence Interval)]= 0.62[0.10, 1.15]; p=0.020). Moreover, those with decreased average familial outness (β[95% C.I.]= −0.66[−0.84, −0.48)]; p<0.001) experienced a significantly greater degree of SGM-related identity concealment since the start of COVID-19.

Additionally, in unadjusted and adjusted regression models (Table 3; only adjusted reported in-text) those who experienced a living rearrangement due to COVID-19 (β[95% C.I.]= 1.56[0.72, 2.41]; p<0.001) and those stably living with parents (β[95% C.I.]= 1.27[0.26, 2.29]; p=0.014) experienced a significantly greater degree of SGM-related familial rejection since the start of COVID-19 compared to those stably living without their parents. Moreover, younger age (β[95% C.I.]= −0.14[−0.19, −0.05)]; p=0.007), being foreign born (β[95% C.I.]= 1.12[0.04, 2.21)]; p=0.042), decreased financial dependence (β[95% C.I.]= −0.36[−0.66, −0.07]; p=0.015), and decreased average familial outness (β[95% C.I.]= −0.71[−1.01, −0.42]; p<0.001) experienced a significantly greater degree of SGM-related familial rejection since the start of COVID-19.

Discussion

Our study hypotheses were partially confirmed. Bivariate analyses indicated that degree of identity concealment and familial rejection since the start of COVID-19 differed among SGM college students who experienced living rearrangement to parents’ homes and who had stable living arrangement with parents in comparison to those with stable living arrangements without parents. However, degree of identity concealment and familial rejection since the start of COVID-19 did not differ between those who experienced living rearrangement to parents’ homes and those who had stable living arrangements with parents. In adjusted regression analyses, we found that those who had experienced a living rearrangement due to COVID-19 suffered from a greater degree of identity concealment and familial rejection since the start of COVID-19 in comparison to those who lived stably without parents. We also found that those who had stable living arrangements with parents suffered from a greater degree of familial rejection since the start of COVID-19 in comparison to those who lived stably without their parents.

Our findings confirm existing minority stress hypotheses, which emphasize that young SGM persons may be prone to experiencing a greater degree of SGM-related stressors as a result of COVID-19,^1,2,9 and are among the first to provide empirical evidence that SGM students who experienced a living rearrangement to their parents homes (identity concealment and familial rejection) or are stably living with their parents (familial rejection only) may be more likely to suffer from increased SGM-related stressors in comparison to those stably not living with their parents. Our findings also revealed that SGM students that experienced a living rearrangement to their parents’ home or were already stably living with their parents had similar degrees of identity concealment and familial rejection since the start of COVID-19. These findings are concerning given that SGM university students have received limited public health attention amid the pandemic.^1,9 Stakeholders must provide attention to the potential secondary minority stress identity-related impacts of the COVID-19 pandemic on SGM young adults and university students, such as familial rejection, which directly impacts their mental health and wellbeing.^27,28 Given existing evidence which suggests that unaffirming familial environments likely constitute a unique vulnerability for some SGM students, particularly during the COVID-19 pandemic,^2,5,9 our findings further support the need for focused research on the ways to support SGM students during this time – particularly for those who experienced a living rearrangement to their parents homes or who were already living with their parents.

Lastly, we found that that SGM college students who are foreign-born are more likely to suffer from an increased degree of familial rejection since the start of COVID-19 compared to those who are U.S.-born. Findings are consistent with previous research regarding the health and mental health vulnerability of SGM immigrants.²⁹ More research is needed to examine the minority stress needs of SGM college students who are foreign-born amid the ongoing COVID-19 pandemic.

Public Health Implications

Mental health and higher education stakeholders must increase the provision of competent and affirming services that address the minority stress and mental health needs of SGM college students who experienced rearrangement to their parents homes, and those who are generally spending more time living with their parents. It is imperative for mental health and higher education stakeholders to leverage existing resources to promote family acceptance and support of SGM youth who are navigating complex and difficult family environments amid COVID-19.^30–33 Further, mental health and higher education services must move beyond traditional tele-health and phone-based services to strengthen the privacy, confidentiality, and safety of SGM young persons, especially those who are living with parents as a result of COVID-19.^2,5,9 Online chat and text-based platforms may be a safe and acceptable solution for SGM young persons living with parents who are seeking support during the pandemic.⁹

Given the established powerful impact of familial rejection on the mental health of SGM young persons,^27,28 it is critical for mental health providers (including in higher education) to identify SGM young persons facing familial rejection (and other minority stressors) in order to mitigate its associated negative mental and physical health ramifications.^1,34 Mental health providers must ensure that they are collecting accurate SGM identity information necessary to identify youth and young adults at risk for identity concealment and familial rejection.^1,35 Providers must also be trained in responding appropriately to SGM youth facing familial rejection, including providing appropriate and affirming care, resources, and referrals.^30–34

Universities can also play an active role in buffering the negative mental health effects of identity concealment and familial rejection among SGM university students by providing resources that promote familial and identity acceptance, and prevent minority stress more broadly.⁵ For instance, universities may increase the capacity of existing SGM-affirmative campus mental health resources (e.g., online and in-person therapy and support groups), and elevate the visibility and dissemination of existing university (e.g., LGBTQ student centers, diversity offices), community (e.g., Trevor Project, National LGBT Help Center, Q Chat Space), and family-related resources for SGM students. Universities could also support SGM students facing familial rejection by offering basic needs support, including emergency and alternative housing options. Lastly, universities could help reduce internalized minority stress (e.g., identity concealment) among SGM students by ensuring that education is delivered in a way that promotes SGM identity and is sensitive toward SGM identity-related concerns.² For instance, instructors can receive training to increase their awareness of SGM-relevant issues (e.g., SGM identity concealment concerns, familial rejection of SGM identities), and to encourage SGM-affirming practices and responses. Instructors can also be encouraged to provide a listing of (SGM-affirming) mental health resources on their syllabi to support SGM and other marginalized students.

Limitations

Firstly, sample size impacted statistical power and our ability to detect significance in multivariate analyses, particularly for identity concealment and those stably living with their parents. Secondly, our study utilized a convenience sampling strategy, which may limit generalizability. Thirdly, some students may have already been planning to go home before university closures, potentially dampening the effect of living rearrangement on our outcomes. Fourthly, as a result of small demographic subgroup sample sizes, our findings do not offer specific estimates for all sexual, gender, racial, and ethnic identity groups. Lastly, we used a retrospective cross-sectional data collection strategy, which limits ability for causal assessment and introduced cognitive bias. Findings should be interpreted carefully and with caution. Despite these limitations, our study provides novel findings and important implications for health and higher education stakeholders to consider for serving SGM students amid COVID-19 and beyond, and helps to set a foundation for future research in this area, particularly longitudinal design studies.

Conclusions

This study found that SGM students who experienced living rearrangement due to COVID-19 or were stably living with their parents suffered from a greater degree of SGM stressors since the start of COVID-19 in comparison to those who lived stably without parents. Our findings emphasize the need for competent and affirming (online) services for SGM students confined to their parents homes (rearranged or stably living), which can mitigate the minority stress impacts of family environments. Through healthcare and higher education reform, findings about the collateral consequences of COVID-19 on SGM students can be leveraged to help address mental health inequities and minority stress amid and beyond the COVID-19 pandemic. Lastly, we implore future researchers to expand upon the COVID-19 experiences and needs of SGM young persons who face multiple marginalization, such as SGM persons of color and/or foreign nationality, those experiencing economic precarity, undocumented status, or unstable housing, and those who are disabled or living in rural locations. Such research is urgently needed to better understand the compounding experiences of minority stress among these highly vulnerable SGM groups, particularly during COVID-19.³⁵