Using a Comprehensive Proximal Stress Model to Predict Alcohol Use

Abstract

Purpose: The gender minority stress model has been proposed by researchers to explain the high rates of substance use found within transgender communities, but its explanatory power has never been directly tested. Further, qualitative research has noted that premedically transitioned and nonbinary individuals are more likely to engage in avoidant coping such as drinking to cope. As such, the current study tested the relationship between proximal stress (comprised of internalized transphobia, anticipated stigma, concealment, and gender dysphoria), drinking to cope, and problematic alcohol use.

Methods: Transgender and nonbinary participants over the age of 18 (N=297) completed an online survey on alcohol use. To test for differences in drinking to cope across gender identity and medical transition, analysis of variance and independent t-tests were conducted. Further, a structural equation model of proximal stress was developed to test whether drinking to cope mediated the relationship between proximal stress and alcohol use.

Results: There were no significant differences across gender identity or medical transition status in severity of drinking to cope. Proximal stress was significantly related to problematic alcohol use and the relationship was explained by an indirect relationship with drinking to cope.

Conclusion: Findings from this study suggest that drinking to cope is an important aspect of high rates of alcohol use found in the transgender community. Implications for clinical practice with transgender individuals are discussed.

Using a Comprehensive Proximal Stress Model to Predict Alcohol Use

The minority stress model describes the unique stressors that impact sexual and gender minorities above and beyond general stressors.^1–3 The model posits that distress is not solely caused by actual prejudice events (distal) but that experiences in a dominant heterosexual/cisgender society also elicit internal (proximal) stressors. For transgender individuals, distal stressors frequently take the form of discrimination across multiple domains, including health care, education, housing, and employment.^4–6 Transgender individuals are also commonly victims of explicit physical and verbal assault and microaggressions.^7,8 Proximal stressors are thoughts and feelings that develop as a result of distal stress, and thus mediate their impact on mental health.⁹ Models of gender minority stress have frequently focused on internalized transphobia (IT), anticipated stigma (AT), and identity concealment as proximal stressors.^3,10–13 Recently, research has proposed and validated that gender dysphoria functions as a proximal stressor for transgender individuals.^14,15

Transgender individuals have disproportionately high rates of alcohol use, with estimates from nonrandom studies reporting hazardous drinking rates to be between 47% and 48% and lifetime alcohol use disorder rates between 11% and 26%.^16–19 The relation between distal stressors and alcohol use is well documented for transgender individuals.^20,21 Increased alcohol use has been linked to physical and sexual assault, gender related discrimination, and public housing discrimination.^22–24

Less is known about the role of proximal stressors in increasing alcohol use among transgender individuals. Research has suggested that for transgender individuals, IT and AT mediate the effects of a having transgender identity on alcohol use; however, existing studies have found mixed results for individual proximal stressors and a full model of proximal stress has yet to be tested.^3,10 A study of gender minority stress and HIV risk behaviors found a relationship between perceived discrimination and alcohol use, but no relationship between IT and alcohol use.²⁵ Additionally, proximal stressors may impact transgender women and men differently. Gonzalez et al. found IT was related to excessive alcohol use for transgender men but not transgender women.¹⁸ However, they did find partial support for some of the proximal stressors of minority stress, together, predicting alcohol use rates. Only one study has investigated gender dysphoria; finding a direct relation between gender dysphoria and alcohol use among transgender women.¹⁸

Drinking to cope represents an avoidant coping skill that may present as a possible explanation for the high rate of alcohol use among transgender individuals.^26–28 Reisner et al. found transmasculine adults report using substances to cope with both enacted and anticipated discrimination in health care settings.²⁹ In addition, 72% of transgender adults in one study noted their main motivation to drink was to cope with negative mood states, a percent far higher than their cisgender counterparts.³⁰ Drinking to cope allows individuals to escape or regulate the negative emotional state caused by proximal stress.³¹ Prior research has documented differences in avoidant coping across both transition status and gender identity. Namely, those with a nonbinary identity or who are in the beginning stages of their gender transition are more likely to utilize avoidant coping skills.^30,32 This is likely because these individuals are not being read by society as their affirmed gender and experience more distal and proximal stressors. However, how individual proximal stressors relate to drinking to cope and gender identity or transition status remain unknown.

This study investigates problematic alcohol use rates among transgender individuals through a comprehensive application of proximal stressors from the gender minority stress model. In alignment with prior research, which established that nonbinary individuals are more likely to use avoidant coping strategies, it was hypothesized that nonbinary individuals would be more likely to endorse drinking to cope than binary transgender (i.e., transmasculine or transfeminine) individuals (hypothesis 1).^30,32 Similarly, as prior research has found that individuals in early stages of medical transition are more likely to utilize avoidant coping, it was hypothesized that those who have not undertaken a medical transition would be more likely to endorse drinking to cope (hypothesis 2).³² Finally, as prior research has shown that the proximal stressors IT and awareness of social stigma together, rather than in isolation, are stronger predictors of alcohol use, it was hypothesized that proximal stress, as a latent variable inclusive of gender dysphoria, would be positively related to problematic alcohol use (hypothesis 3a).¹⁸ Further, because transgender individuals have reported utilizing drinking to cope with negative mood states, it was predicted that drinking to cope would mediate the relationship between proximal stress and problematic alcohol use (hypothesis 3b).³⁰

Method

Procedure

Before data collection, this study was approved by the Towson University Institutional Review Board. Participants were recruited from announcements, posted to social networking sites, calling for participants who identified as transgender and were at least 18 years old. Participants most frequently accessed the survey via Reddit (62.3%) and Facebook (27.3%). The survey began with an informed consent document followed by a demographic questionnaire, measure items, and concluded with a prompt thanking the participants and provided an opportunity to leave suggestions to improve future studies. As part of the demographics, participants provided their gender as a write-in response in addition to selected fixed, static, identity choices. The fixed responses for gender identity were utilized in the this article for statistical analysis. Participants received no financial incentive to participate in the study.

Participants

Three hundred completed the survey; however, 3 participants under the age of 18 participated in the survey and were removed before analysis. This left a sample size of 297 participants who identified as transfeminine (n=128), transmasculine (n=110), nonbinary (n=49), and agender (n=10).^* Participants ranged in age from 18 to 67 (M=28, SD=9.42) and represented 42 states, Washington, D.C. and 24 countries. The sample had limited racial/ethnic diversity with 84.0% of the sample identifying as White. Table 1 includes further demographic information.

Table 1.

Participant Demographics

	Total (N=297)
Age, mean (SD)	28.04 (9.42)
Gender identity (%)
Nonbinary	19.6
Transfeminine	43.3
Transmasculine	37.0
Ethnicity (%)
American Indian/Alaskan Native	0.3
Asian/Asian American	1.7
Biracial/Multiracial	3.7
Black/African American	0.7
Hispanic/Latinx	5.0
White	84.0
No answer	1.0
Other	3.0
Sexual identity (%)
Asexual	5.7
Bisexual	21.7
Fluid	0.7
Gay	6.7
Heterosexual	10.7
Lesbian	17.3
Pansexual	18.3
Queer	15.7
Other	3.3
Socioeconomic status
Working class	38.3
Middle class	43.4
Upper class	4.1
Other	5.8
Don't know	5.8
No answer	2.7
Education level (%)
High school degree/GED	46.7
College degree	38.0
Graduate degree	10.0
Doctorate/terminal degree	5.3

Measures

Proximal stress

The measures for AS, IT, and concealment were taken from the Gender Minority Stress and Resilience Measure.¹¹ Responses for all three scales are recorded on a 5-point scale from 0 (strongly disagree) to 4 (strongly agree), items are summed with higher scores indicating greater endorsement. All three scales showed high internal consistency for the current study (α between 0.86 and 0.92).

Gender dysphoria was measured using the Gender Congruence (GC) subscale of the Gender Congruence and Life Satisfaction Scale.³³ The GC subscale contains 17 items and is composed of 4 subscales: genitalia, chest, other secondary sex characteristics, and social gender role recognition. Items are rated on a 5-point scale from 1 (always) to 5 (never) and are averaged to obtain subscale scores. The GC subscale was reversed scored where low scores reflected lower severity of gender dysphoria. The GC subscale had high internal consistency (α=0.84) in this sample.

Alcohol use and coping

Problematic alcohol use was assessed using the Alcohol Use Disorders Identification Test (AUDIT).³⁴ The scale contains 10 questions that assess rates of alcohol consumption and negative behavioral outcomes. Responses are rated on various scales that range between 0 and 5 (items 1 and 2) or 0–4 (items 3–10). Scale items are summed, with scores of eight or higher indicating problematic alcohol use. Participants scores were then converted into a dichotomous variable with scores below 8 coded as 0 and scores above 8 coded as 1. The scale showed high internal consistency (α=0.83) in this sample.

The coping subscale of the Drinking Motive Questionnaire Revised Short Form was used to measure drinking to cope.³⁵ The subscale contains three items that assess the utilization of drinking to cope with negative mood states. Responses are coded on a 3-point scale from 1 (never) to 3 (almost always), items are averaged with higher scores indicating higher levels of drinking to cope. The scale showed high internal consistency (α=0.89) in this sample.

Medical transition status

To assess medical transition status, participants were asked whether they “have pursued any medical transition steps to affirm their gender?” To avoid legitimizing certain medical transition steps over others, we allowed participants to define for themselves what constituted a medical transition step. Those who answered “yes” were coded as 1 and those answered “no” or “not interested” were coded as 0.

Data analysis

The data were analyzed with SPSS version 26 and Amos version 23. First, the data were screened for univariate and multivariate outliers, in addition to skewness and kurtosis. Next, for gender identity and medical transition, analysis of variance (ANOVA) and independent t-tests were conducted to test for significant differences in drinking to cope (hypothesis 1 and 2). Finally, a structural equation model (SEM) was used to assess the relation between proximal stress and problematic alcohol use (hypothesis 3a), and to test whether this relationship was mediated by drinking to cope (hypothesis 3b). It should be noted that due to the cross-sectional nature of this study, a true test of mediation could not be performed; however, analysis of direct and indirect paths allowed us to test the total relationship between the three variables.³⁶ Model fit criteria of comparative fit index (CFI) >0.90, root mean square error of approximation (RMSEA) <0.10, and standardized root mean square residual (SRMR) <0.10 was utilized to assess fit.^37–39

Results

Data screen

When evaluating the assumptions of multivariate normality and linearity, 10 univariate outliers (3 in drinking to cope, 3 in AS, and 4 in AUDIT) were removed. The data were additionally evaluated for multivariate outliers using Mahalanobis distances that indicated that two were multivariate outliers (p<0.001) and were removed from analysis. This left a final sample size of 295. After deletion of outliers, three scales were still significantly skewed (skewness index <3, kurtosis index <10); therefore, multiple indicators of model fit were assessed.⁴⁰ Missing values in all scales did not exceed 1.5%, were distributed randomly across measures and cases, and were addressed with regression imputation.

Descriptive statistics

The average scores across gender identity for proximal stressors and outcome measures can be found in Table 2. ANOVAs were conducted to test for differences across gender identity for each of the proximal stressors. Significant group differences were found for AT, concealment, and gender dysphoria, but not IT. Ad hoc pairwise comparisons were conducted yielding the following significant differences: transfeminine individuals endorsed higher levels of AT than transmasculine individuals; transfeminine individuals endorsed higher levels of gender identity concealment than nonbinary individuals; and transfeminine individuals endorsed higher levels of gender dysphoria than transmasculine and nonbinary individuals. All other pairwise comparisons were not significant. Table 2 includes findings specific to each of the four subscales of gender dysphoria. Finally, there were no significant differences in levels of drinking to cope or problematic alcohol use across gender identity or medical transition status when measured continuously or dichotomously (Tables 2 and 3).

Table 2.

Proximal Stressors, Alcohol Use, and Drinking to Cope Across Gender Identity

	Total	Transfeminine	Transmasculine	Nonbinary	F (2, 292)	η2p
	(N=295)	(n=127)	(n=109)	(n=49)
	M (SD)	M (SD)	M (SD)	M (SD)
Proximal stressors
Internalized transphobia	15.44 (8.18)	16.70 (8.77)	14.75 (7.32)	14.00 (8.14)	2.84	0.02
Concealment	13.01 (5.26)	13.62 (5.36)	13.21 (4.88)	11.30 (5.40)	4.22*,a	0.03
Anticipated stigma	24.75 (7.15)	26.16 (7.04)	23.48 (7.22)	24.08 (6.84)	3.97*,b	0.03
Gender dysphoria	3.16 (.74)	3.42 (0.71)	3.04 (0.75)	2.84 (0.60)	16.03***,c	0.10
Social	3.13 (0.98)	3.35 (0.92)	2.72 (1.04)	3.43 (0.69)	17.09***,d	0.11
Genitalia	2.94 (0.92)	3.20 (0.88)	2.95 (0.91)	2.34 (0.74)	19.91***,e	0.12
Chest	3.30 (1.25)	3.24 (1.11)	3.58 (1.38)	2.96 (1.20)	5.06**,f	0.04
Other secondary sex characteristics	3.41 (1.23)	4.25 (0.86)	2.77 (1.12)	2.80 (1.01)	77.25***,g	0.35
Outcome measures
Drinking to cope	1.50 (0.65)	1.52 (0.69)	1.50 (0.63)	1.46 (0.64)	0.243	0.00
AUDIT score	3.99 (4.85)	3.92 (5.22)	4.18 (4.84)	3.98 (4.27)	5.52	0.00
	N (%)	n (%)	n (%)	n (%)	χ2 (p)	V
Problematic alcohol use	49 (16.6)	18 (14.2)	19 (17.4)	12 (20.3)	1.19 (.552)	-

Bold values indicate significance.

^*p<0.05, ^**p<0.01, ^***p<0.001.

^aTransfeminine>nonbinary.

^bTransfeminine>transmasculine.

^cTransfeminine>transmasculine and nonbinary.

^dTransfeminine and nonbinary>transmasculine.

^eTransfeminine and trans masculine>nonbinary.

^fTransmasculine>nonbinary.

^gTransfeminine>transmasculine and nonbinary.

AUDIT, Alcohol Use Disorders Identification Test.

Table 3.

Proximal Stressors, Alcohol Use, and Drinking to Cope Across Medical Transition Status

	Total	Medically transitioned	Not medically transitioned	t (df)	p	d
	(N=295)	(n=200)	(n=95)
	M (SD)	M (SD)	M (SD)
Proximal stressors
Internalized transphobia	15.44 (8.18)	14.34 (8.27)	17.77 (7.52)	3.43 (293)	0.001	0.43
Anticipated stigma	24.75 (7.15)	24.03 (7.52)	26.09 (6.10)	2.50 (223.64)	0.013	0.31
Concealment	13.01 (5.26)	13.11 (5.43)	12.82 (4.82)	−0.443 (293)	0.658	—
Gender dysphoria	3.16 (0.74)	3.11 (0.79)	3.28 (0.63)	1.95 (225.56)	0.052	—
Social	3.13 (0.98)	2.92 (1.01)	3.59 (0.74)	6.44 (243.90)	<0.001	0.80
Genitalia	2.94 (0.92)	3.03 (0.95)	2.74 (0.83)	−2.53 (290)	0.012	0.32
Chest	3.30 (1.25)	3.20 (1.33)	3.51 (1.05)	2.14 (231.64)	0.034	0.27
Other secondary sex characteristics	3.41 (1.23)	3.35 (1.28)	3.57 (1.12)	1.51 (208.70)	0.133	—
Outcome measures
Drinking to cope	1.50 (0.65)	1.49 (0.65)	1.50 (0.66)	0.08 (293)	0.933	—
AUDIT score	3.99 (4.85)	4.11 (4.91)	3.76 (4.75)	−0.57 (293)	0.567	—
	N (%)	n (%)	n (%)	χ2	p	Φ
Problematic alcohol use	49 (16.6)	33 (16.5)	16 (16.8)	0.01	0.941	-

Bold values indicate significance.

Differences across problematic alcohol use by gender identity

Differences in levels of proximal stressors and drinking to cope by problematic alcohol use and gender identity were assessed through independent samples t-tests, where the sample was split by gender identity, with the results presented in Table 4. For the total sample, participants who endorsed problematic alcohol use also endorsed higher levels of IT (d=0.42) and AT (d=0.36; Table 5). Similarly, within the sub-sample of nonbinary participants, those who endorsed problematic alcohol use also endorsed higher levels of AT (d=0.66; Table 4). Additionally, within the sub-sample of transfeminine participants, those who endorsed problematic alcohol use also endorsed higher levels of social gender dysphoria (d=0.57; Table 4). Across all gender identities, individuals who endorsed problematic alcohol use were significantly more likely to endorse greater levels of drinking to cope (d ranging between 1.25 and 1.71).

Table 4.

Proximal Stressors and Drinking to Cope Across Problematic Alcohol Use by Gender Identity

	Transfeminine					Transmasculine					Nonbinary
	(n=127)					(n=109)					(n=59)
	Problematic drinking	Not problematic	t (df)	p	d	Problematic drinking	Not problematic	t (df)	p	d	Problematic drinking	Not problematic	t (df)	p	d
	(n=18)	(n=109)				(n=19)	(n=90)				(n=12)	(n=47)
	M (SD)	M (SD)				M (SD)	M (SD)				M (SD)	M (SD)
Proximal stressors
Internalized transphobia	20.28 (8.86)	16.11 (8.66)	1.89 (125)	0.062	—	16.95 (7.53)	14.29 (7.23)	1.45 (107)	0.151	—	17.33 (7.88)	13.15 (8.07)	1.61 (57)	0.113	—
Anticipated stigma	27.06 (7.05)	25.83 (7.14)	0.68 (125)	0.501	—	26.05 (5.94)	22.95 (7.30)	1.73 (107)	0.086	—	27.58 (6.23)	23.19 (6.76)	2.04 (57)	0.046	0.66
Concealment	15.44 (3.94)	13.34 (5.47)	1.57 (125)	0.120	—	14.11 (4.19)	13.02 (5.01)	0.89 (107)	0.382	—	12.82 (5.26)	10.91 (5.43)	1.09 (57)	0.280	—
Gender dysphoria	3.67 (0.66)	3.38 (0.72)	1.60 (125)	0.112	—	3.17 (0.53)	3.01 (0.79)	1.07 (37.02)	0.291	—	3.07 (0.57)	2.78 (0.60)	1.50 (57)	0.141	—
Social	3.79 (0.86)	3.27 (0.92)	2.22 (125)	0.028	0.57	2.69 (0.68)	2.73 (1.10)	−0.21 (41.17)	0.834	—	3.39 (0.72)	3.60 (0.56)	−0.097 (57)	0.338	—
Genitalia	3.17 (0.87)	3.21 (0.89)	−0.19 (122)	0.851	—	3.16 (0.75)	2.91 (0.94)	1.10 (107)	0.276	—	2.50 (0.76)	2.30 (0.74)	0.87 (57)	0.387	—
Chest	3.60 (1.04)	3.18 (1.12)	1.51 (123)	0.134	—	3.82 (1.11)	3.53 (1.43)	0.94 (28.14)	0.358	—	3.19 (1.09)	2.90 (1.23)	0.75 (57)	0.459	—
Other secondarya	4.46 (0.72)	4.21 (0.88)	1.13 (125)	0.259	—	2.90 (1.11)	2.74 (1.13)	0.55 (105)	0.581	—	3.31 (0.95)	2.67 (0.99)	1.98 (57)	0.052	—
Outcome measures
Drinking to cope	2.35 (0.71)	1.38 (0.59)	6.315 (125)	<0.001	1.61	2.09 (0.55)	1.37 (0.57)	4.96 (107)	<0.001	1.25	2.22 (0.61)	1.24 (0.42)	5.28 (57)	<0.001	1.71

Bold values indicate significance.

^aOther secondary sex characteristics.

Table 5.

Proximal Stressors and Drinking to Cope Across Problematic Alcohol Use by Medical Transition Status

	Total					Medically transitioned					Not medically transitioned
	(N=295)					(n=200)					(n=95)
	Problematic drinking	Not problematic	t (df)	p	d	Problematic drinking	Not problematic	t (df)	p	d	Problematic drinking	Not problematic	t (df)	p	d
	(n=49)	(n=246)				(n=33)	(n=167)				(n=16)	(n=79)
	M (SD)	M (SD)				M (SD)	M (SD)				M (SD)	M (SD)
Proximal stressors
Internalized transphobia	18.27 (8.11)	14.88 (8.10)	2.67 (293)	0.008	0.42	17.48 (8.29)	13.71 (8.15)	2.42 (198)	0.016	0.46	19.88 (7.72)	17.34 (7.46)	1.23 (93)	0.221	—
Anticipated stigma	26.80 (6.33)	24.27 (7.23)	2.27 (293)	0.024	0.36	25.76 (6.64)	23.69 (7.65)	1.45 (198)	0.149	—	28.94 (5.20)	25.51 (6.13)	2.09 (93)	0.040	0.57
Concealment	14.28 (4.41)	12.76 (5.36)	1.87 (293)	0.063	—	14.82 (3.84)	12.77 (5.64)	2.57 (62.84)	0.013	0.49	13.18 (5.37)	12.74 (4.73)	0.33 (93)	0.745	—
Gender dysphoria	3.33 (.64)	3.13 (.76)	1.71 (293)	0.088	—	3.36 (.66)	3.06 (.80)	2.01 (198)	0.046	0.38	3.27 (.60)	3.28 (.64)	0.07 (93)	0.942	—
Social	3.31 (0.87)	3.10 (0.99)	1.44 (293)	0.151	—	3.17 (0.94)	2.87 (1.01)	1.61 (198)	0.109	—	3.61 (0.64)	3.58 (0.76)	0.14 (93)	0.890	—
Genitalia	3.00 (0.83)	2.92 (0.94)	0.52 (290)	0.602	—	3.11 (0.83)	3.01 (0.97)	053 (195)	0.599	—	2.78 (0.81)	2.73 (0.84)	0.18 (93)	0.855	—
Chest	3.58 (1.09)	3.25 (1.27)	1.85 (73.49)	0.069	—	3.76 (1.11)	3.09 (1.34)	2.94 (48.65)	0.005	0.59	3.23 (0.96)	3.56 (1.06)	−1.17 (93)	0.245	—
Other secondarya	3.57 (1.16)	3.39 (1.24)	0.97 (291)	0.332	—	3.48 (1.24)	3.32 (1.28)	0.65 (196)	0.515	—	3.77 (0.97)	3.23 (1.15)	0.79 (93)	0.431	—
Outcome measures
Drinking to cope	2.22 (.63)	1.35 (.55)	9.01 (63.83)	<0.001	1.41	2.13 (.63)	1.37 (.58)	6.86 (198)	<0.001	1.31	2.40 (.60)	1.32 (.51)	6.73 (19.55)	<0.001	1.85

Bold values indicate significance.

^aOther secondary sex characteristic.

Differences across problematic alcohol use by medical transition status

Differences in levels of proximal stressors and drinking to cope by problematic alcohol use were assessed through independent samples t-tests, where the sample was split by medical transition status, with the results presented in Table 5. For participants who had medically transitioned, problematic alcohol use was related to significantly higher levels of IT (d=0.46), concealment (d=0.49), total gender dysphoria (d=0.38), and chest gender dysphoria (d=0.59). For participants who had not transitioned, problematic alcohol use was related to significantly higher levels of AT (d=0.57). Across both groups, problematic alcohol use was related to significantly higher levels of drinking to cope (d ranging between 1.31 and 1.85).

Proximal stressors and problematic alcohol use

Correlations between individual proximal stressors and problematic alcohol use are presented in Table 6. All variables were significantly positively related to each other; however, the relationship between proximal stressors and problematic alcohol use were weak.

Table 6.

Correlations Between Proximal Stressors and AUDIT Scores

	1	2	3	4
1. Internalized transphobia
2. Concealment	0.50**
3. Anticipated stigma	0.46**	0.47**
4. Gender dysphoria	0.45**	0.47**	0.41**
5. Problematic drinking	0.17**	0.13*	0.15*	0.16**

^*p<0.05, ^**p<0.01.

Structural equation model

The proposed model of proximal stress was tested and validated in a previous study.¹⁵ This model of proximal stress was then analyzed via SEM to test the relationship between proximal stress, drinking to cope, and problematic alcohol use. This model showed acceptable fit to the data, CFI=0.89, RMSEA=0.06 (90% CI: 0.06–0.06), and SRMR=0.08.^37–39 As shown in Figure 1, proximal stress was a significant predictor of problematic alcohol use behaviors (β=0.22, p=0.001) and drinking to cope (β=0.31, p<0.001). When both predictors were included in the model, drinking to cope predicted alcohol use behaviors (β=0.55, p<0.001), but proximal stress did not (β=0.04, p=0.448), indicating that the relation between proximal stress and problematic alcohol use was explained by the indirect relationship with drinking to cope.

Fig. 1.

SEM of drinking to cope mediating the effects of proximal stress on problematic alcohol use, with coefficients presented in standardized form. ***p<0.001.

Discussion

Transgender individuals are at an elevated risk for problematic alcohol use,¹⁷ and within this sample, 16.6% of participants meet criteria for problematic alcohol use as determined by WHO criteria.³⁴ As hypothesized, the present results support that proximal stress is positively related to problematic alcohol use, and this effect was explained by an indirect relation with drinking to cope. However, contrary to our hypotheses, nonbinary individuals and those who had not undertaken a medical transition did not endorse significantly greater severity of drinking to cope than binary or medically transitioned individuals.

Proximal stress and alcohol use

Although prior studies have tested individual components of proximal stress and their relation to alcohol use, the present study is the first to test a comprehensive model of proximal stress inclusive of IT, AT, identity concealment, and gender dysphoria. Our results demonstrated that all four proximal stressors were independently related to problematic alcohol use, but these positive associations were weak. Our findings additionally support the notion that when all components are taken together, proximal stress predicts alcohol use, such that individuals who experienced elevated levels of proximal stress were more likely to endorse problematic drinking. These findings mirror past research showing that proximal stress is a stronger predictor of alcohol use than individual stressors alone.¹⁸ Overall, these results support the proposed impact of gender minority stress raised by Hendricks and Testa by showing that individuals who experience elevated gender related stress do engage in negative health risk behaviors, specifically problematic alcohol use.³ It is also important to note that gender dysphoria was found to have a significant positive relationship with problematic alcohol use. While gender dysphoria has only recently been proposed to function as a proximal stressor in models of gender minority stress, this study provides supplemental support for the potential role of gender dysphoria as a mediator between distal stress and negative health outcomes.^14,15

Drinking to cope

As proximal stress represents the internal processing of external experiences of prejudice, it was hypothesized that drinking to cope would mediate the relation between proximal stress and problematic alcohol use. This hypothesis was supported, as drinking to cope was found to indirectly affect the relation, such that individuals who experienced high levels of proximal stress were more likely to report drinking to cope and to have increased levels of problematic alcohol use. This indicates that alcohol is used as a source of coping to deal with the experience of gender related stress.

Participants who met criteria for problematic drinking were more likely to endorse high levels of drinking to cope. This result was consistent when comparing participants across gender identity or medical transition status. Thus, it appears that individuals who drink to cope consume greater amounts of alcohol and endorse more behavioral consequences for their drinking. However, contrary to our hypotheses, nonbinary individuals and individuals who had not transitioned, were not significantly more likely to report drinking to cope than any other group. This finding stands in stark contrast to the assumptions set forth by prior coping research.^30,32 However, our findings may be more reflective of a lack of drinking in our sample, rather than a lack of avoidant coping utilization. As not all transgender individuals engage in alcohol use, and only 16.6% of our sample reached the level of problematic use, it is possible that our participants were engaging in other avoidant coping skills that were not assessed as part of this study. As such, future research should investigate the extent to which drinking to cope is utilized in comparison to other avoidant coping skills.

Limitations

The present study extends our current understanding of proximal stress and substance use by proposing a comprehensive model of proximal stress; however, it is not without limitations. We utilized a convenience sample of online participants that can disproportionately sample White, educated, and middle-class participants.⁴¹ The current sample is not an exception and results should be interpreted in light of this limitation. However, online sampling allowed us to recruit a sample of transgender individuals who endorsed diverse gender and sexual identities and may not have been accessible otherwise.⁴² Additionally, our limited sample size did not allow for the model of proximal stress to be examined by gender identity. Future research should aim to replicate the findings of the relationship between proximal stress, drinking to cope, and alcohol use across gender identities. Furthermore, the measure of gender dysphoria utilized did not allow us to measure body-based gender dysphoria as a construct separate from social gender dysphoria. As body-based dysphoria can impact individuals differently than social gender dysphoria, it will be important for future research to investigate whether these aspects have differential effects on drinking to cope and alcohol use.^14,43 Finally, because outliers were removed from the data set, the findings may not generalize to all transgender individuals, especially those with high levels of alcohol consumption. Considering this, and the low level of problematic alcohol consumption within the current sample, it may be especially important for researchers to replicate these findings within a sample of transgender individuals who endorse high levels of alcohol consumption and related problems.

Implications

These results have important implications for understanding problematic alcohol use among transgender individuals. By utilizing a comprehensive proximal stress model, this study showed that individuals who experience elevated levels of proximal stress are more likely to seek out alcohol to cope with the distress. Further, individuals who were utilizing alcohol to cope were more likely to have negative alcohol related behavioral outcomes as result. These results highlight a need for the development of substance use interventions that target positive coping skills. Recent developments within resilience-based approaches to minority stress can be applied to these interventions.⁴⁴ Because prior studies have found that social support is negatively correlated with avoidant coping, it may be especially important for transgender individuals with problematic alcohol use to seek out social networks when developing more positive coping strategies.³² These networks provide a space where transgender individuals can discuss their experiences with proximal stress and provide modeling of positive ways to cope with minority stress.⁴⁴

Abbreviations Used

ANOVA

analysis of variance

AS

anticipated stigma

CFI

comparative fit index

GC

Gender Congruence

IT

internalized transphobia

AUDIT

Alcohol Use Disorders Identification Test

RMSEA

root mean square error of approximation

SEM

structural equation model

SRMR

standardized root mean square residual

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this research.

Cite this article as: Lindley L, Bauerband L, Galupo MP (2021) Using a comprehensive proximal stress model to predict alcohol use, Transgender Health 6:3, 164–174, DOI: 10.1089/trgh.2020.0042.