LGBTQ+ Outpatients Present to Eating Disorder Treatment Earlier and With More Severe Depressive Symptoms Than Cisgender Heterosexual Peers

Abstract

Community evidence indicates high eating disorder (ED) and comorbid symptom severity among LGBTQ+ compared to cisgender heterosexual (CH) individuals. Little is known about such disparities in ED treatment samples, especially in outpatient treatment. We aimed to descriptively characterize and investigate baseline group differences in symptom severity between LGBTQ+ and CH ED outpatients at treatment intake. Data from 60 (22.3%) LGBTQ+ and 209 (77.7%) CH ED outpatients were used to examine: (1) demographic and diagnostic differences; 2) differences in ED, depressive, and emotion dysregulation symptoms. Objectives were tested using Fisher-Freeman-Halton exact and independent samples t-tests, and analyses of covariance adjusted for age and diagnosis, respectively. Most LGBTQ+ outpatients were bisexual (55.2%), and 6.5% identified as transgender and non-binary. LGBTQ+ outpatients presented to treatment at younger ages (Mean Difference [MD] = −3.39, p = .016) and reported more severe depressive symptoms (MD = 5.73, p = .004) than CH patients, but endorsed similar ED symptom and emotion dysregulation severity. Groups did not differ in other demographic or diagnostic characteristics. LGBTQ+ individuals may develop more severe depression and similarly severe EDs at earlier ages but seek outpatient care sooner than CH peers. Managing depressive symptoms may be particularly important for LGBTQ+ ED patients.

Introduction

Eating disorders (EDs) are multifactorial illnesses that have complex etiological underpinnings, significant medical consequences, and high mortality rates (Voderholzer et al., 2020). Community data indicate more severe ED symptomatology and higher incidence of ED diagnoses for LGBTQ+ individuals (e.g., lesbian, gay, bisexual, transgender, and nonbinary people) than their cisgender heterosexual (CH) peers (Grammer et al., 2021; Kamody et al., 2020; Parker & Harriger, 2020). Specifically, LGBTQ+ individuals report higher odds of anorexia nervosa, bulimia nervosa, and binge-eating disorder, and more severe restricting, purging, binge-eating, and shape and weight concern symptoms (Grammer et al., 2021; Kamody et al., 2020; Parker & Harriger, 2020). Researchers are just beginning to examine whether these inequities appear in clinical LGBTQ+ samples. Findings from partial hospital (Donahue et al., 2020) and intensive outpatient through residential settings indicate higher global ED severity among LGBTQ+ than CH patients at admission (Brewerton et al., 2022; Mensinger et al., 2020), with one study finding higher preoccupation and restriction, fasting, self-induced vomiting, and shape and weight concerns among subgroups of sexual minority compared to heterosexual women (Murray et al., 2021).

Beyond ED symptom differences, community and clinical LGBTQ+ samples report worse psychological health than CH individuals. For example, two studies found higher depressive, emotion dysregulation, and trauma symptoms among LGBTQ+ relative to CH patients in partial hospital and residential ED treatment (Brewerton et al., 2022; Donahue et al., 2020). Depression and emotion dysregulation are known correlates of ED symptomatology (Prefit et al., 2019; Puccio et al., 2016), and higher emotion dysregulation in LGBTQ+ individuals may partially explain community differences in ED symptomatology (Gillikin et al., 2021). However, findings from Gillikin and colleagues (2021) are cross-sectional and thus temporal processes predicting differential ED severity remain unclear.

Further, it is unknown whether disparities in ED and comorbid symptoms observed between LGBTQ+ and CH patients in higher levels of care extend to outpatients. This is concerning for several reasons. Outpatient treatment is considered the first-line defense for EDs (Anderson et al., 2017). This may be the only care level LGBTQ+ individuals are able to access due to more limited economic resources and health coverage compared to CH peers, and the significant financial burden associated with higher levels of ED treatment (Charlton et al., 2018; Streatfeild et al., 2021). Outpatient treatment typically focuses on modification of eating behaviors and core psychopathology such as body image distress (Atwood & Friedman, 2020) but there may be nuanced and unique maintenance factors for LGBTQ+ patients that current treatments are not designed to accommodate. Ultimately, research examining efficacy of evidence-based ED treatments for LGBTQ+ patients does not exist.

This paucity of treatment data, coupled with a lack of clarity around clinical profiles of LGBTQ+ ED outpatients, may result in the application of treatments that are not ideally suited to their needs. To this end, LGBTQ+ individuals across the age range report discriminatory, stigmatizing, and otherwise negative experiences with mental healthcare, including ED treatment (Cusack et al., 2022; Duffy et al., 2016). Similarly, despite improvements in awareness of ED disparities between LGBTQ+ and CH patients, clinicians describe feeling unprepared to meet the needs of LGBTQ+ patients in ED treatment (Ferrucci et al., 2023). It is thus critical to understand ED and comorbid symptom profiles of LGBTQ+ patients across the care continuum, especially at the outpatient level, to better inform future research on affirming ED care for LGBTQ+ populations.

This study aimed to characterize demographic and diagnostic profiles, and explore baseline symptom differences, among LGBTQ+ and CH ED outpatients upon presentation to treatment. We tested: (1) group differences in demographic and clinical variables (e.g., age at intake, race and ethnicity, ED diagnosis); and 2) differences in frequency and severity of ED (i.e., preoccupation and restriction, dietary restraint, shape and weight concerns, fasting, binge eating, self-induced vomiting, laxative misuse, and compulsive exercise), depressive, and emotion dysregulation symptoms between LGBTQ+ and CH ED outpatients. We hypothesized that LGBTQ+ individuals would report more frequent and severe depressive, emotion dysregulation, and ED symptoms than CH outpatients.

Materials and Methods

Participants

This study utilized retrospective data from a clinical outcomes study at an outpatient ED clinic. Of 403 patients who completed an intake assessment between April 2017 and December 2022, 270 (67%) consented to have their data used for research purposes. One participant who did not identify their sexuality was excluded. The final sample included 269 adolescents and adults. Two-hundred-nine (77.7%) patients self-identified as CH and 60 (22.3%) as LGBTQ+. See Table 1 for demographic and diagnostic data. LGBTQ+ patients identified their sexualities as heterosexual (n = 1, 1.7%), lesbian (n = 8, 13.8%), gay (n = 4, 6.9%), asexual (n = 5, 8.6%), and “other” (n = 8, 13.8%) including one pansexual participant, with two participants who did not disclose. LGBTQ+ patients identified their genders as female (n = 48, 78.7%), male (n = 7, 11.5%), nonbinary (n = 3, 4.9%), transgender (n = 1, 1.6%), and “other” (n = 2, 3.3%) without a text response. CH patients included 187 (89.5%) females and 22 (10.5%) males. Patients were included if they met Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria for a primary ED diagnosis (American Psychiatric Association, 2013). Patients received an ED diagnosis following a semi-structured clinical interview administered by a licensed clinical psychologist, master’s- or doctoral-level psychology trainees supervised by the licensed clinical psychologist, and/or licensed clinical social workers.

Table 1.

Sample Demographic and Diagnostic Characteristics.

	LGBTQ+ (n = 60, 22.3%)	CH (n = 209, 77.7%)
Variable	M (SD), range	M (SD), range
Age at intake	22.36 (9.52), 10–72a	25.95 (14.14), 9–81b
BMI	25.16 (10.79), 14.61–57.80a	24.64 (10.40), 11.90–63.84a
Age of ED onset	14.05 (4.67), 1–30a	15.50 (7.90), 1–51a
Sexuality	n (%)	n (%)
Heterosexual	1 (1.7)	209 (100.0)
Lesbian	8 (13.8)	-
Gay	4 (6.9)	-
Bisexual	32 (55.2)	-
Asexual	5 (8.6)	-
“Other”	8 (13.8)	-
Not reported	2 (3.3)
Gender	n (column %, row %)	n (column %, row %)
Female	47 (78.3, 20.1)	187 (89.5, 79.9)
Male	7 (11.7, 24.1)	22 (10.5, 75.9)
Non-binary	3 (5.0, 100.0)	-
Transgender	1 (1.7, 100.0)	-
“Other”	2 (3.3, 100.0)	-
Race/Ethnicity
White	34 (56.7, 20.6)	131 (62.7, 79.4)
Hispanic or Latinx	12 (20.0, 30.8)	27 (12.9, 69.2)
Black	4 (6.7, 19.0)	17 (8.1, 81.0)
Asian	4 (6.7, 18.2)	18 (8.6, 81.8)
Native Hawaiian or Pacific Islander	0 (0, 0)	1 (0.5, 100.0)
American Indian or Alaska Native	0 (0, 0)	1 (0.5, 100.0)
Multiple races	6 (10.0, 35.3)	11 (5.3, 64.7)
Not reported	-	3 (1.4, 75.0)
ED Diagnosis
Anorexia Nervosa	15 (25.0, 20.0)	60 (28.7, 80.0)
Bulimia Nervosa	10 (16.7, 22.2)	35 (16.7, 77.8)
Binge-Eating Disorder	12 (20.0, 22.6)	41 (19.6, 77.4)
Avoidant Restrictive Food Intake Disorder	3 (5.0, 15.0)	17 (8.1, 85.0)
OSFED Atypical Anorexia Nervosa	12 (20.0, 25.0)	36 (17.2, 75.0)
OSFED subthreshold Bulimia Nervosa	1 (1.7, 20.0)	4 (1.9, 80.0)
OSFED subthreshold Binge-Eating Disorder	2 (3.3, 28.6)	5 (2.4, 71.4)
OSFED Purging Disorder	4 (6.7, 40.0)	6 (2.9, 60.0)
Unspecified feeding or ED	1 (1.7, 16.7)	5 (2.4, 83.3)

Note. Different superscripts denote significant group difference in age. Participants who reported their sexuality as “Other” and provided a text response include 1 pansexual participant.

CH, cisgender heterosexual; ED, eating disorder; BMI, body mass index; OSFED, Other Specified Feeding or Eating Disorder.

Procedure

Study procedures were approved by the University of Chicago Institutional Review Board. All patients who presented for intake were provided information about the study and invited to participate. Adult patients provided written consent. Adolescent patients provided written assent; parents/guardians provided informed consent on their behalf. After obtaining consent, clinic staff measured height and weight, respectively, via stadiometer and digital scale. Patients completed self-report measures via a secure iPad-based system or a paper questionnaire packet prior to clinical interview assessments.

Measures

Eating Disorder Examination Questionnaire (EDE-Q)

Frequency and severity of ED symptoms were measured using the EDE-Q (Fairburn & Beglin, 1994). Although there is evidence for high reliability and validity of the EDE-Q in adolescents and adults with EDs, evidence from sexual minority samples suggests an alternate four-factor structure, previously examined in non-sexual-minority samples, that may fit this construct better (Klimek et al., 2021). Thus, we administered 11-item shape and weight concern, five-item preoccupation and restriction (e.g., excessive and impairing thoughts about food or body, food avoidance), and three-item dietary restraint (e.g., deliberate attempts to restrict eating) subscales. A fourth three-item eating shame subscale was not used due to poor reliability (α’s = 0.59–0.64) across subsamples. Certain ED behaviors (e.g., binge eating) were also examined at the item level given prior item-level differences observed in clinical samples of LGBTQ+ people (e.g., Murray et al., 2021). Cronbach’s alpha values for the three utilized subscales were 0.82–0.94 across subsamples.

Center for Epidemiologic Studies–Depression Scale Revised (CESD-R)

Depressive symptom severity was measured using the child/adolescent and adult versions of the CESD-R (Eaton et al., 2004), which exhibits strong psychometric properties, including high internal consistency, strong factor loadings, and convergent validity with negative affect (Haroz et al., 2014; Van Dam & Earleywine, 2011). The CESD-R has demonstrated psychometric invariance in adult sexual minority samples (Gomez & McLaren, 2017; Miedema et al., 2019). In the current sample, internal consistency was high; α = .91 for adults, α = .90 for adolescents, and α = .90–.92 in subsamples.

Difficulties in Emotion Regulation Scale (DERS)

Emotion dysregulation was assessed using the DERS, which provides a total score that sums subscales assessing the following domains: (1) lack of clarity, (2) non-acceptance of emotional responses, (3) impulse control difficulties when upset, (4) difficulties engaging in goal-directed behavior when upset, and (5) limited access to emotion regulation strategies (Gratz & Roemer, 2004). Six items assessing lack of emotional awareness were removed because they do not fit the factor structure, and the DERS demonstrates high internal consistency, good test-retest reliability, and adequate construct and predictive validity without them (Hallion et al., 2018). We are unaware of psychometric evaluation of the DERS in LGBTQ+ samples although internal consistency was high in prior research with transgender (Pinciotti et al., 2022) and sexual minority samples (Donahue et al., 2020). In this study, internal consistency was high; α = .94 in the total sample, α = .92–.94 in subsamples.

Statistical Analyses

Analyses were conducted in SPSS v29.0 (Armonk, NY: IBM Corp.). Fisher-Freeman-Halton exact tests were used to evaluate associations between race and ethnicity, ED diagnosis, and group (i.e., LGBTQ+ or CH). Independent samples t-tests were used to evaluate group differences in age at intake, BMI, and self-reported age of ED onset. Cohen’s d effect sizes were computed for t-tests (small effect: d = 0.2, medium effect: d = 0.5, large effect: d = 0.8). One-way analyses of covariance (ANCOVAs), adjusting for age and ED diagnosis, were used to examine group differences in ED, depressive, and emotion dysregulation symptoms. Although groups did not differ by ED diagnosis, diagnosis was included as a covariate in ANCOVAs because certain ED symptoms (e.g., self-induced vomiting) may be less relevant to certain diagnoses (e.g., binge-eating disorder). Partial eta squared effect sizes were computed for ANCOVAs (small effect:η_p² = 0.01, medium effect: η_p² = 0.06, large effect: η_p² = 0.14).

LGBTQ+ patients were aggregated into one group in effort to maximize power given the smaller sample size of these patients. However, to ensure that specific populations were not driving the pattern of results, we ran a series of additional mean comparison analyses to assess whether LGBTQ+ subgroups differed on variables of interest. For example, bisexual participants were contrasted with non-bisexual participants given that these individuals comprised a majority of the LGBTQ+ sample. Similarly, the 12 (20.0%) LGBTQ+ patients who did not identify as women were contrasted with the 48 (80.0%) who did. No significant differences were observed on any variables including age at intake, or depressive, emotion dysregulation, and ED symptoms. Further, we re-examined group difference analyses between LGBTQ+ and CH patients after removing the four transgender and non-binary patients. Results did not meaningfully change; therefore, we present only those results including the full, aggregated sample. Although this study is preliminary in nature, due to the smaller sample size of LGBTQ+ patients and number of tests, Benjamini-Hochberg correction was applied across analyses (Benjamini & Hochberg, 1995).

Results

At a corrected significance threshold of p = .017, LGBTQ+ patients presented to treatment at younger ages than CH patients, Mean Difference (MD) = −3.39, t(267) = −2.16, p = .016, d = 0.28, 95%CI: −6.48, −0.29. Groups did not differ by BMI, t(261) = 0.44, p = .330, self-reported age of ED onset, t(246) = −1.72, p = .043, race and ethnicity (Fisher-Freeman-Halton exact = 5.08, p = .662), or ED diagnosis (Fisher-Freeman-Halton exact = 3.45, p = .918).

LGBTQ+ patients reported more severe depressive symptoms than CH patients (MD = 5.73, p = .004, η_p² = .030), which remained significant after Benjamini-Hochberg correction (p < .005; Table 2). Groups did not differ on emotion dysregulation, preoccupation and restriction, dietary restraint, shape and weight concerns, or fasting, binge eating, self-induced vomiting, laxative use, and compulsive exercise frequency (Table 2).

Table 2.

Analyses of Covariance Results for Eating Disorder and Comorbid Symptoms.

	LGBTQ+ (n = 60)	CH (n = 209)
	M (SD)	M (SD)	F	p	η p 2
CESD-R	29.90 (13.98)	24.17 (12.07)	7.70	.004*	.030
DERS	83.72 (23.37)	81.40 (27.78)	0.01	.961	.000
Preoccupation and restriction	2.80 (1.71)	2.53 (1.75)	0.96	.329	.004
Cognitive restraint	3.37 (2.41)	3.11 (2.22)	0.43	.514	.002
Shape and weight concerns	3.93 (1.61)	3.68 (1.80)	1.43	.232	.006
Fasting	1.16 (1.53)	1.23 (1.86)	0.09	.763	.000
Binge eating	7.34 (9.39)	7.24 (9.53)	0.29	.592	.001
Self-induced vomiting	1.73 (5.65)	3.67 (12.10)	1.26	.264	.005
Laxative use	0.98 (3.65)	1.28 (8.15)	0.04	.841	.000
Compulsive exercise	3.47 (6.52)	3.71 (7.61)	0.04	.852	.000

Note. Analyses adjusted for age and eating disorder diagnosis. CH, cisgender heterosexual; CESD-R, Center for Epidemiological Studies Depression-Revised; DERS, Difficulties in Emotion Regulation Scale.

^*adjusted p < .005.

Discussion

This study extends prior research by examining whether known disparities in ED symptomatology and comorbid psychopathology among LGBTQ+ and CH ED patients apply to outpatients. In our sample, 22.6% identified as LGBTQ+ and these individuals did not differ in ED diagnosis or self-reported age of ED onset compared to CH patients.

No ED symptom differences were observed between LGBTQ+ and CH ED outpatients. Recent research in clinical samples suggests higher global ED severity among LGBTQ+ patients and differences in specific symptoms (e.g., self-induced vomiting) between subgroups of sexual minority and heterosexual women (Donahue et al., 2020; Mensinger et al., 2020; Murray et al., 2021). These studies utilized larger samples (i.e., between 398 and 3,120 cases) and we may have been underpowered to capture such differences. Additionally, these studies each examined samples in higher levels of care for eating disorders (e.g., partial hospitalization). It is possible that outpatient and higher-level-of-care samples differ in presentation as a function of care access, patient appropriateness for each level of care, or differences in the structure of treatments or these settings themselves. However, direct comparison of patient characteristics across the levels of care is needed before such conclusions can be drawn. Finally, it may also be that ED symptoms are more similar among LGBTQ+ and CH patients at the point they present to ED treatment or that treatment-seeking and community samples differ in presentation. Symptom differences between groups in higher levels of care tend to be small in magnitude (e.g., Murray et al, 2021) and less pronounced than disparities observed in community samples (Parker & Harriger, 2020). Thus, ED symptomatology differences may begin to converge among those patients who can and do access treatment. There is some evidence to support this type of treatment and community sample discrepancy in Black women with binge-eating disorder, such that eating pathology in a clinical sample of Black women differed significantly from a community sample of Black women, but not a clinical sample of white women (Grilo et al., 2005). Our preliminary findings and the small number of prior studies suggest that these disparities require further examination to determine the extent to which ED severity differs upon presentation to treatment across the care continuum.

Unexpectedly, LGBTQ+ and CH patients also endorsed similar levels of emotion dysregulation, which is inconsistent with community and limited treatment data (Donahue et al., 2020). Our results should be interpreted cautiously given the small sample size of LGBTQ+ patients but suggest that these individuals may not consistently report higher emotion dysregulation than CH individuals upon presenting to ED treatment. However, it is important to consider that even if emotion dysregulation itself does not differ between sexual and gender diverse groups, it may serve a distinct function for LGBTQ+ people. For example, in response to minority stressors such as stigma, LGBTQ+ individuals may require more emotion regulatory processes than CH individuals and may be more prone to impulsive and maladaptive coping strategies (Gilliken et al., 2021). Similarly, a large body of evidence has shown higher incidence and severity of psychological distress including depression for LGBTQ+ individuals beginning in youth (Lucassen et al., 2017), and emotion regulatory processes could partly explain disparities in this distress and maladaptive coping (Gillikin et al., 2021; Hatzenbuehler, 2009). Ultimately, we did not examine stress factors and mediation analysis was avoided given the cross-sectional design of this study, but future research should consider the ways in which these factors interact to predict similar ED severity at younger ages for LGBTQ+ ED patients.

To this end, it is concerning that LGBTQ+ patients endorsed more severe depressive symptoms with similarly severe ED symptoms at younger ages than CH patients. Effect sizes observed for these differences were small, which may reflect that the subsamples shared many characteristics or relate to the smaller sample size of LGBTQ+ patients. However, findings are consistent with higher-level-of-care data and suggest that more severe ED-related comorbid depression may develop at younger ages for LGBTQ+ than CH individuals (Donahue et al., 2020). Further, because groups did not differ in self-reported age of ED onset, findings suggest that LGBTQ+ individuals may develop more severe ED symptoms that necessitate treatment more quickly than CH individuals. Findings may also indicate that LGBTQ+ individuals seek outpatient ED care sooner than CH peers, which is notable as early detection and intervention can increase ED care utilization, improve prognostic and treatment outcomes, and reduce morbidity and mortality (Kalindjian et al., 2022). However, it is important to emphasize that LGBTQ+ individuals face delays in accessing ED care as well as substantial barriers to diagnosis, referral, and treatment (Mensinger et al., 2020).

In sum, present findings contribute to a small but growing body of evidence that LGBTQ+ individuals are likely to present to ED treatment, potentially at younger ages, with unique needs such as attention to more severe depressive symptoms. Thus, ED clinicians may need to modify existing treatments to better address LGBTQ+ patients’ needs. For example, cognitive behavioral and transdiagnostic treatment models may need to apply a minority stress framework to better conceptualize more severe depression and emotion regulatory processes as precipitating and maintaining factors in LGBTQ+ patients’ ED etiology (Labarta & Emelianchik-Key, 2022). Dialectical behavior therapy may be well-suited to target depression and emotion dysregulation in the context of ED symptoms, and increase distress tolerance resources to cope with unique stressors; family-based treatment could incorporate identity development and familial support around identity when working with affirming families (Calzo et al., 2020).

In the absence of treatment evidence, however, it is more broadly necessary that ED clinicians are sensitive to the limitations of existing ED treatments and LGBTQ+ patients’ experiences of those treatments as at times incongruent with sexual and gender identity needs (Cusack et al., 2022; Duffy et al., 2016). Outpatient mental health clinicians have a unique opportunity to set the stage for continued ED treatment and are implored to incorporate best practices for LGBTQ+ patients into their care. These may include but are not limited to: deeper understanding of the unique minority stress, affective, and coping factors in ED development; knowledge of interpersonal, familial, and community dynamics; awareness of LGBTQ+ resources; and incorporation of LGBTQ+ identity development, bodily autonomy, and empowerment in treatment (Pachankis et al., 2021). At a structural level, ED researchers and clinicians alike are stakeholders in increasing care accessibility, supporting the education and cultural awareness of treatment program staff, and cultivating treatment spaces that are safe and respectful for LGBTQ+ patients (Calzo et al., 2020).

Strengths and Limitations

This study contributes to very limited data regarding ED and comorbid psychopathology among clinical LGBTQ+ populations by providing the first comparative characterization of LGBTQ+ and CH outpatients with EDs. Findings may help clarify distinct ED correlates among LGBTQ+ and CH patients across the ED care continuum and inform prospective examination of ED risk and severity. However, findings should be interpreted cautiously given the small sample size of LGBTQ+ patients in this study. Additionally, examination of differences over the course of treatment was unfortunately not feasible due to large percentages of missingness (e.g., 80%) across follow-up time points. Future studies should expand upon the present study by examining ED and comorbid psychopathology among LGBTQ+ individuals across ED treatment course.

Use of a collapsed LGBTQ+ group in this study prevents a more nuanced understanding of clinical profiles among LGBTQ+ subgroups. Although subgroups of the LGBTQ+ patients did not appear to be driving our pattern of results, ED risk does differ among LGBTQ+ subgroups (Grammer et al., 2021) and disparities in ED symptoms have been observed among sexual minority subgroups in a clinical sample of women (Murray et al., 2021). Growing evidence suggests that bisexual people, particularly bisexual women, are at greatest risk to develop EDs, potentially as a function of dual stigmatization by both CH and other sexual minority peers (Brewster & Moradi, 2010; Hazzard et al., 2020). It is notable that bisexual people comprised more than half of our LGBTQ+ subsample, indicating that these individuals may be in greatest need of or most likely to present to ED treatment. Identification of etiological factors relevant to distinct LGBTQ+ groups with EDs is needed to inform their treatment. Although more detailed LGBTQ+ subgroup analyses in future studies with larger clinical samples is important, this study provides needed foundational information at the outpatient level to guide future research.

Similarly, although the study sample was slightly more racially and ethnically diverse than prior treatment studies, participants were predominantly self-identified white women. Like for LGBTQ+ individuals, research on EDs among Black, Indigenous, and People of Color is sparse. ED symptoms do differ across racial and ethnic groups, and non-white patients in ED treatment may feel isolated, tokenized, or misunderstood (Watson, 2020). Future studies should investigate clinical profiles at intersections of different identities given evidence of differential ED severity among multiply marginalized individuals (Austin et al., 2013).

Finally, our study only included assessment of identity categories without other indicators such as sexual attraction or sexual behavior. Recent evidence suggests that assessment of identity alone may contribute to discrepant findings related to ED psychopathology in the literature (O’Flynn et al., 2022). Future research may benefit from incorporating multiple assessments of identity, which appears to produce more convergent findings (O’Flynn et al., 2022). Nevertheless, the ability to collect large enough treatment samples to disaggregate LGBTQ+ populations or conduct intersectional research—especially in non-privatized and smaller evidence-based programs—is likely contingent upon advocacy efforts to increase accessibility to ED treatment for historically excluded populations. Such efforts must include greater attention to these populations in ED literature. Finally, as LGBTQ+ individuals experience unique care barriers, such as more limited health coverage than CH individuals (Charlton et al., 2018), it is unclear if clinical samples are misrepresentative of LGBTQ+ individuals more broadly. Future studies of LGBTQ+ populations in ED treatment contexts should account for factors such as insurance more directly to better understand ways in which current evidence may be skewed as a function of treatment access.

Conclusion

The present study reifies important differences in psychological health among LGBTQ+ ED patients and is the first, to our knowledge, to do so in an outpatient setting. Findings underscore a need for prospective studies with LGBTQ+ populations to better understand how depressive and emotion dysregulation symptoms may interact to predict ED severity at younger ages for LGBTQ+ individuals. Examining associations across the ED care continuum is critical. Like prior studies (e.g., Donahue et al., 2020), LGBTQ+ individuals were overrepresented in this sample (22.6%) relative to population estimates (5.6%; Jones, 2021). In other words, health systems and ED clinics at all care levels are likely to treat LGBTQ+ patients with EDs and therefore need to be aware of their specific needs.

Clinical Implications.

• LGBTQ+ outpatients presented to eating disorder (ED) treatment at younger ages than cisgender heterosexual (CH) peers.

• LGBTQ+ outpatients endorsed more severe depressive symptoms than CH peers but similarly severe EDs and emotion dysregulation.

• Prospective studies are needed to understand the role of more severe depressive symptoms in ED severity and treatment needs.

• ED treatments may need to be modified to better address LGBTQ+ individuals’ needs.

Data Availability Statement:

The data that support the findings of this study are available upon reasonable request from the University of Chicago Eating Disorders Program. The data are not publicly available due to privacy or ethical restrictions.