Internalized Stigma in Patients with Eosinophilic Gastrointestinal Disorders

Abstract

The aim of the current study is to evaluate internalized stigma in individuals diagnosed with an eosinophilic gastrointestinal disorder (EGID) and its impact on psychosocial and health-related outcomes. The final study sample consisted of 149 patients with a self-reported EGID diagnosis for at least six months. Participants completed measures evaluating internalized stigma, disease-specific quality of life, emotional distress (anxiety, depression) and answered questions regarding healthcare utilization. Overall, increased internalized stigma was associated with decreased disease-specific quality of life, and increased anxiety and depression. In addition, participants with greater overall internalized stigma felt that treatments were less effective, and the internalized stigma subscales of alienation and discrimination were associated with increased outpatient visits and endoscopies, respectively. Providers working with EGID patients should assess for signs of internalized stigma, such as social withdrawal and alienation. Psychogastroenterology services that deliver evidence-based psychological interventions may reduce some of the negative impacts of internalized stigma.

Introduction

Eosinophilic Gastrointestinal Disorders (EGIDs) are a group of chronic gastrointestinal (GI) disorders marked by infiltration of eosinophilia to one or more segments of the GI tract (Dellon et al., 2013; Liacouras et al., 2011; Rothenberg, 2004). There are four EGID subtypes, defined by the location affected: the most common, eosinophilic esophagitis (EoE), eosinophilic gastritis (EG), eosinophilic gastroenteritis (EGE), and eosinophilic colitis (EC) (Gonsalves, Furuta, & Atkins, 2016; Rothenberg, 2004). EGIDs often cause physiological changes to the associated part of the GI tract (e.g. inflammation, ulcers, strictures, stenosis), resulting in symptoms of difficulty swallowing, edema, malabsorption, food impaction, bowel obstruction, nausea, vomiting, pain, and diarrhea (Gonsalves et al., 2016; Rothenberg, 2004). Overall, the histopathology of EGIDs is not well understood. However, genetic and environmental factors have been implicated, including food allergy (Rothenberg, 2004). Treatment options are limited, but consist of pharmacological and dietary options, such as steroids, acid suppression, and elimination and elemental diets (Dellon et al., 2013; Gonsalves, 2018; Liacouras et al., 2011; Rothenberg, 2004). Finally, while still relatively new and rare, EGIDs are increasing in prevalence and incidence, underscoring the importance of evaluating their impact on the individuals living with these conditions (Dellon, Jensen, Martin, Shaheen, & Kappelman, 2014; Jensen, Martin, Kappelman, & Dellon, 2016).

Understanding the psychosocial consequences of living with one or more of these disorders is important to deliver adequate patient-centered care. Prior research identifies negative impacts of EGIDs on health-related quality of life (HRQOL) (Bedell et al., 2018; Taft, Kern, Keefer, Burstein, & Hirano, 2011; Taft, Kern, Kwiatek, et al., 2011). Factors such as the relative newness and lack of awareness of EGIDs, changes in diet and eating habits, and the subsequent impact on social and psychological functioning contributes to the decrease in HRQOL (Bedell et al., 2018; Taft, Kern, Keefer, et al., 2011). In addition, individuals with EGIDs report several challenges and unmet needs across medical, healthcare, social and emotional domains of functioning (Hiremath et al., 2018). Given the rare nature of EGIDs coupled with these psychosocial consequences, stigma is important to consider when working with this population. Our group recently evaluated the impact of perceived stigma, defined as the subjective awareness of stigmatizing attitudes, beliefs, or behaviors others may engage in because a person has a particular trait, in individuals with EGIDs and its impact on HRQOL and other psychosocial functioning (Guadagnoli, Taft, & Keefer, 2017). Increased perceived stigma was predictive of greater depression, anxiety and illness impact, as well as decreased HRQOL. In addition, perceived stigma was the strongest predictor of each of the outcome variables, underscoring its significant influence on individuals with EGIDs.

Another domain within the construct of stigma is internalized stigma, which occurs when an individual incorporates stigmatizing attitudes into their self-identity and believes negative attitudes or stereotypes regarding their illness are true and apply specifically to them (Corrigan, 1998). It is comprised of four domains; alienation, stereotype endorsement, discrimination experiences, and social withdrawal (Boyd, Adler, Otilingam, & Peters, 2014; Ritsher, Otilingam, & Grajales, 2003). Internalized stigma has been evaluated across several chronic illness populations (Huggett et al., 2018; Waugh, Byrne, & Nicholas, 2014; Zeng et al., 2018) and in gastrointestinal illnesses such as inflammatory bowel disease (IBD) (Taft, Ballou, & Keefer, 2013; Taft & Keefer, 2016) and irritable bowel syndrome (IBS) (Taft, Riehl, Dowjotas, & Keefer, 2014). However, to date, no research evaluates internalized stigma in patients with EGIDs and the impact on HRQOL, psychological functioning, and other patient reported outcomes.

Our study’s primary aim is to evaluate internalized stigma in a cohort of patients with EGIDs. Secondary aims are to assess associations of internalized stigma with: 1) health related quality of life, 2) treatment adherence, 3) healthcare utilization, and 4) emotional distress (anxiety, depression).

Methods

Adults aged 18-70 diagnosed with an EGID for at least six months were recruited from a university-based outpatient gastroenterology practice, patient advocacy groups (American partnership for Eosinophilic GI disorders (APFED), Campaign Urging Research for Eosinophilic Disease (CURED) foundation), and social media (Facebook, Twitter). After providing informed consent, participants completed screening questionnaires to assess for exclusion criteria and to verify EGID diagnosis. In an effort to address confounding stigma from other diagnoses, patients were excluded from participation if they endorsed a diagnosis of another GI illness (e.g. IBS or IBD) or a serious mental illness (e.g. schizophrenia, schizoaffective disorder, severe bipolar, depression with psychotic disorders, and personality disorders), as internalized stigma has been established these disorders (Boyd et al., 2014; Huggett et al., 2018; Taft & Keefer, 2016; Taft et al., 2014). This study was approved by the Institutional Review Board at Northwestern University. Eligible participants completed the following questionnaires:

Demographic Information:

Age, gender, marital status, race, ethnicity, education level.

Clinical Information:

EGID diagnosis (“What is your EGID diagnosis?” with options: esophagitis, gastritis, gastroenteritis, colitis, more than one EGID, I have not been diagnosed with EGID), past misdiagnosis (yes/no), diagnosis duration (years), symptom duration prior to diagnosis (years), current number of EGID medications, and current dietary treatment (yes/no). Participants completed a series of self-rated scales assessing disease severity (0-10, 10 = very severe), disease control over the past month (0-10, 10 = extremely well), treatment efficacy over the past month (0-10, 10 = extremely well).

Healthcare Utilization:

Number of outpatient visits, endoscopies, emergency department visits (each in the past year) due to their EGID, and indicated (yes/no) if they have ever sought treatment from a dietician or mental health professional (psychologist or counselor).

Treatment Adherence:

Medication adherence over the past week (0-100, 100 = exactly as prescribed), and EGID dietary adherence over the past week (0-100, 100 = exactly as prescribed).

The Internalized Stigma In Mental Illness Scale (Adapted for EGID):

The ISMI is a validated self-report questionnaire the evaluates subjective experience of internalized stigma (Ritsher et al., 2003). The ISMI includes 29 items on a 4-point Likert scale (1 = Strongly Disagree, 4 = Strongly Agree). Higher scores indicate increased internalized stigma. Internalized stigma is assessed overall via a global score and across the following four subscales: Alienation (“I feel out of place in the world because of my [EGID]”), Stereotype Endorsement (“People with an [EGID] cannot live a good, rewarding life”), Discrimination Experience (“Nobody would be interested in getting close to me because I have an [EGID]”), Social Withdrawal (“I avoid getting close to those who don’t have an [EGID] to avoid rejection”). The optional fifth subscale “Stigma Resistance” was not used in this study. The item “Mentally ill people tend to be violent” was removed from the questionnaire because it was not applicable. In addition, the words “Mental Illness” were replaced with “EGID” for all remaining items. The ISMI has demonstrated good reliability (Cronbach α = .90), test-retest reliability (r=.92), and construct validity (Ritsher et al., 2003). The ISMI has been used for other chronic conditions in this manner (Taft et al., 2013; Taft & Keefer, 2016; Taft et al., 2014).

Eosinophilic Esophagitis Quality of Life Scale – Adult:

The EoE-QOL-A is a validated self-report questionnaire designed to evaluate disease-specific HRQOL in EoE patients (Taft, Kern, Kwiatek, et al., 2011). HRQOL is assessed across the following five domains: Eating/diet impact, social impact, emotional impact, disease anxiety, and choking anxiety. The EoE-QOL-A includes 37 questions on a 5-point Likert scale (0 = “Extremely”, 4 = “Not at All”). Higher scores indicate better HRQOL. For individuals with non-EoE EGID, the EoE-QOL-A was modified so that ‘EoE’ was changed to ‘EGID’ for each question of relevance. Internal consistency for the modified version was excellent (Cronbach α = 0.98, Guttman split-half reliability = 0.96). Further, our group used the modified EoE-QOL-A in a previous study to evaluate HRQOL in EGIDs (Guadagnoli et al., 2017).

NIH PROMIS short-form: Emotional-Distress (Anxiety and Depression).

The NIH PROMIS Anxiety and Depression scales were used to evaluate levels of emotional distress in the study sample (Pilkonis et al., 2011). Both scales include 8 items and are ranked on a 5-point Likert Scale (0 = Never, 4 = Always). The anxiety scale evaluates self-reported fear, anxious misery, hyperarousal, and somatic symptoms related to arousal. The depression scale evaluates self-reported negative mood, views, of self, social cognition, and decreased positive affect and engagement. PROMIS scales demonstrate excellent internal consistency (Anxiety: Cronbach α = 0.79; Depression: Cronbach α = 0.83) (Pilkonis et al., 2011).

Statistical Analyses:

Analyses were conducted using IBM SPSS v24 for Macintosh operating systems (Chicago, IL). Ineligible participants were removed from the sample prior to analyses. Descriptive statistics, including skewness and kurtosis, were evaluated to assess the need for nonparametric tests. Mean, median, standard deviation, and range were calculated for continuous variables and percentages were calculated for categorical variables. One-way ANOVA evaluated differences between recruitment source for total score and subscales and between the four EGID diagnoses (EoE, EG, EGE, EC) on demographic and clinical variables to determine if disease groups should be pooled together or treated separately for analyses.

Independent samples t-tests evaluated differences between demographic and clinical variables between internalized stigma, HRQOL, anxiety, and depression scores. Pearson’s and Spearman’s correlations assessed for relationships between internalized stigma (total, alienation, stereotype endorsement, discrimination, social withdrawal) and continuous demographic and clinical variables. In addition, correlations evaluated associations between the six internalized stigma scores and disease-specific HRQOL, anxiety, and depression.

Three separate hierarchical multiple regression analyses were conducted to evaluate internalized stigma as a predictor for the following psychosocial outcome variables: disease-specific HRQOL, anxiety, and depression. Three additional hierarchical multiple regression analyses evaluated internalized stigma as a predictor for the following healthcare utilization outcomes: number of outpatient visits, endoscopies, and emergency department visits. For all analyses, clinical or demographic variables that were significantly correlated at P < .10 with internalized stigma were entered into the model at Step 1 to control for any potential confounding effects. Adjusted R², R² change, standardized beta weight, and standard error are reported.

To control for potential Type 1 error, P was adjusted to <. 01 for statistical significance based on Bonferroni correction for between-subjects comparisons. To be consistent, for correlations we also set P < .01 based on a Bonferroni correction for 42 regression analyses per table rather than pooling all correlations for correction as to avoid Type II error. For regression analyses, statistical significance was set to P < .05 due to the exploratory nature of these analyses.

Results

In total, 210 participants visited the online questionnaire portal and 209 consented to participate in the study. Sixty participants were excluded for incomplete data or not meeting study inclusion criteria resulting in a final sample size of 149 (72% completion rate). Demographic and clinical data is presented in Table 1. The majority of the sample was diagnosed with EoE (77%), followed by EGE (6%), and equal representation of EG and EC (1.3%). No significant differences existed between disease groups for demographic or clinic data. Therefore, all of the subjects were pooled together for analyses. Internalized stigma was significantly correlated with several clinical variables (Table 2).

Table 1.

Demographic and Clinical information

	% (N)
Gender
Male	22.4% (33)
Female	77.6% (114)
Married	56.1% (83)
Caucasian/White	93.8% (137)
Non-Hispanic	96.8% (121)
College Educated	66.7% (88)
EGID Diagnosis
EoE	77.2% (115)
EG	1.3% (2)
EGE	6.0% (9)
EC	1.3% (2)
>1 EGID	14.1% (21)
Recruitment Source
Clinic	9.4% (14)
Patient Organization	10.1% (15)
Social Media	65.7% (98)
Other	14.8% (22)
Misdiagnosed in Past	71.0% (103)
Current Dietary Treatment	53.4% (78)
Seen Dietitian for EGID	45.2% (66)
Seen Mental Health Professional for EGID	15.8% (23)
	Mean (SD)
Age	37.3 (10.3)
EGID Severity (0-10 scale)	6.69 (2.1)
EGID Control Past Month (0-10)	6.73 (2.8)
EGID Treatment Efficacy Past Month (0-10)	6.51 (2.8)
EGID Medication Adherence Past week (0-100)	79.15 (31.4)
EGID Diet Adherence Past Week (0-100)	72.75 (33.4)
	Median (Range)
Years Since Symptom Start	7 (0-45)
Years Since Diagnosis	3 (0-30)
Healthcare Utilization (Past Year)
Outpatient Appointments	5 (0-50)
Endoscopies	1 (0-19)
Emergency Department Visits	0 (0-12)
Number of EGID Medications	2 (0-14)

Table 2.

Correlations between Internalized Stigma and Healthcare Utilization, Treatment Adherence, and Disease Severity

	IS Total	Alienation	Stereotype	Discrimination	Withdrawal
Outpatient Visits	0.18	0.23*	0.15	0.21	0.19
Endoscopies	0.17	0.15	0.14	0.27*	0.16
ED Visits	0.04	0.03	0.06	0.14	0.02
# Medications	0.06	0.05	0.12	0.15	0.11
Disease Severity	0.30*	0.24*	0.21	0.30*	0.29*
Disease Control	−0.20	−0.19	−0.21	−0.26*	−0.20
Treatment Efficacy	−0.29*	−0.27*	−0.31*	−0.27*	−0.20
Medication Adherence	0.02	0.00	−0.09	0.00	0.05
Diet Adherence	0.09	0.06	0.07	0.06	0.15

^*p < .01;

Note: significance was set at the p<.01 level to account for potential Type 1 error.

Differences in several study variables existed by gender, dietary treatment, and treatment by a mental health professional. Overall, women (78% of the sample) reported significantly higher internalized stigma, depression/anxiety, and HRQOL compared to men (all p < .01). When evaluated by internalized stigma subscale (e.g. alienation, stereotype endorsement, discrimination, and withdrawal), these differences remained for all subscales except stereotype endorsement. Just over half (53%) of the sample endorsed currently using dietary treatment and those on dietary treatment reported higher internalized stigma (p < .01) and HRQOL, including the dietary and emotional subscales (all p < .01) compared to those not currently on dietary treatment. Approximately 16% of the sample reported seeing a mental health professional specifically for issues related to their EGID. Those who saw a mental health professional reported higher internalized stigma across all scales except for stereotype endorsement (p < .01), and reported increased emotional HRQOL (p< .001) compared to those who have not seen a mental health professional. No differences existed on any other demographic or clinical variables including marital status, misdiagnosis, dietician use, or recruitment source. Due to the small sample of non-white and Hispanic participants, differences between the sample on race or ethnicity could not be assessed.

Internalized Stigma Associations with Disease and Treatment

Internalized stigma was associated with several clinical and healthcare utilization variables (Table 2). Patients with greater internalized stigma reported increased disease severity (r = .30, p < .01) and decreased treatment efficacy (r = −.29, p < .01). These associations remained significant across the internalized stigma subscales, aside from stereotype endorsement and withdrawal, which were not significantly correlated with disease severity and treatment efficacy, respectively. Discrimination was the only internalized stigma variable significantly associated with number of endoscopies (r = .27, p < .01) and perceived disease control (r = −0.26, p <.01), while alienation was the only internalized stigma variable significantly associated with number of outpatient visits (r = 0.23, p <.01).

Internalized Stigma and Psychosocial Function

Several significant correlations exist between internalized stigma and psychosocial domains, such as HRQOL, anxiety, and depression (Table 3). Overall, internalized stigma was significantly associated with decreased HRQOL (r = −.51, p < .01). When evaluated by HRQOL subscale, overall internalized stigma was greatest for the emotional impact (r = −.65, P < .01), followed eating/diet impact (r = −.43, p < .01), and disease anxiety (r = −.40, p < .01). Internalized stigma was also significantly associated with increased depression (r = .65, p < .01) and anxiety (r = .56; p < .01). Subscale analyses found all of these relationships remained significant. While relationships between overall internalized stigma and the HRQOL subscales fell outside of the Type 1 corrected significance cutoff of .01, internalized stigma subscale analyses did reveal several significant associations. Patients with increased HRQOL related to choking anxiety experienced decrease alienation (r = −0.29), discrimination (r = −0.23), and withdrawal (r = −23, all p < .01). In addition, decreased social HRQOL was associated with increased alienation (r = −0.25, p < .01).

Table 3.

Correlations between Internalized Stigma and disease-specific HRQOL, Depression, and Anxiety

	IS Total	Alienation	Stereotype	Discrimination	Withdrawal
HRQOL Total	−0.51*	−0.60*	−0.41*	−0.46*	−0.53*
Eating/Diet Impact	−0.43*	−0.52*	−0.33*	−0.39*	−0.48*
Social Impact	−0.18	−0.25*	−0.11	−0.21	−0.21
Emotional Impact	−0.65*	−0.72*	−0.55*	−0.54*	−0.65*
Disease Anxiety	−0.40*	−0.46*	−0.35*	−0.38*	−0.40*
Choking Anxiety	−0.20	−0.29*	−0.19	−0.23*	−0.23*
Depression	0.65*	0.70*	0.52*	0.51*	0.59*
Anxiety	0.56*	0.58*	0.46*	0.50*	0.56*

^*p < .01;

Note: significance was set at the p<.01 level to account for potential Type 1 error.

Internalized Stigma as a Predictor of HRQOL and Emotional Distress

Hierarchical linear regression analyses evaluated internalized stigma as a predictor for several psychosocial variables such as HRQOL, anxiety, and depression (Table 4). Findings demonstrated alienation as a significant predictor for HRQOL, anxiety, and depression, when controlling for disease severity and symptom control. No other internalized stigma subscales, including stereotype endorsement, discrimination, or withdrawal, remained significant predictors.

Table 4.

Hierarchical Linear Regression Analyses for Internalized Stigma as predictor of QOL, Anxiety, and Depression when controlling for EGID severity and control

	R2 adjusted	R2 change	B	SE	P
HRQOL Total
Model 1	0.12	0.13			< .001
EGID Severity			−0.25	0.99	.004
Sx Control			0.23	0.72	.006
Model 2	0.37	0.26			< .001
Alienation			−0.40	4.38	.002
Stereotype			0.03	5.97	.812
Discrimination			0.02	4.63	.870
Withdrawal			−0.21	4.25	.120
Anxiety
Model 1	0.03	0.04			.081
EGID Severity			0.08	0.30	.364
Sx Control			−0.17	0.22	.059
Model 2	0.34	0.33			< .001
Alienation			0.40	1.29	.003
Stereotype			0.01	1.77	.994
Discrimination			0.09	1.36	.464
Withdrawal			0.16	1.29	.252
Depression
Model 1	0.03	0.04			.067
EGID Severity			0.06	0.34	.524
Sx Control			−0.19	0.24	.036
Model 2	0.49	0.47			< .001
Alienation			0.69	1.27	< .001
Stereotype			0.01	1.80	.984
Discrimination			0.02	1.33	.984
Withdrawal			0.43	1.23	.665

Internalized Stigma as a Predictor of Healthcare Utilization

Additional regression analyses assessed internalized stigma as a predictor for healthcare utilization, including number of outpatient visits, endoscopies, and emergency department visits (Table 5). For this analysis, a log10 conversion was done for all 3 criterion variables due to non-normal data. When controlling for disease severity and symptom control, discrimination experiences predicted the number of endoscopies, while social withdrawal predicted the number of emergency department visits. There were no significant predictors for number of outpatient visits.

Table 5.

Hierarchical Linear Regression Analyses for Internalized Stigma as predictor of Healthcare Utilization when controlling for EGID severity and control

	R2 adjusted	R2 change	B	SE	P
Outpatient Visits
Model 1	0.05	0.07			.018
EGID Severity			−0.10	0.02	.310
Sx Control			−0.24	0.01	.012
Model 2	0.04	0.02			.681
Alienation			0.07	0.09	.672
Stereotype			0.05	0.13	.719
Discrimination			0.12	0.10	.453
Withdrawal			−0.07	0.09	.663
Endoscopies
Model 1	0.03	0.05			.105
EGID Severity			0.20	0.02	.046
Sx Control			0.10	0.01	.311
Model 2	0.15	0.15			.002
Alienation			−0.25	0.06	.114
Stereotype			0.09	0.09	.515
Discrimination			0.56	0.07	.001
Withdrawal			−0.13	0.06	.146
ED Visits
Model 1	0.01	0.07			.305
EGID Severity			−0.03	0.02	.846
Sx Control			−0.27	0.02	.128
Model 2	0.08	0.17			.194
Alienation			0.01	0.14	.969
Stereotype			0.16	0.18	.545
Discrimination			0.57	0.14	.072
Withdrawal			−0.85	0.15	.024

Discussion

The current study sought to evaluate internalized stigma in patients with EGIDs, a relatively rare group of diseases that affect areas of the GI tract, and its impact on several clinical and psychosocial outcomes. Overall, internalized stigma was associated with increased disease severity. The relationship between internalized stigma and these perceived disease factors has been studied in other digestive illnesses, including IBS and IBD. Although EGIDs differ from IBS and IBD in a number of ways (e.g. pathophysiological origin, treatments), they share common symptoms depending on the disease type including nausea, vomiting, diarrhea, and inflammation (e.g. IBD) (Quigley, 2016). Further, they are all concealable, chronic, and incurable diseases of the GI tract, which increases the susceptibility to stigma (Guadagnoli et al., 2017; E. E. Jones, 1984; Quigley, 2016). Internalized stigma is associated with increased symptom severity in patients with IBS (Taft et al., 2014), and IBD patients in remission endorse less internalized stigma and engage in more stigma-resistant behaviors than IBD patients not in remission (Taft et al., 2013).

Increased healthcare utilization is an important outcome to consider in EGID research. As a whole, EGID management may involve several endoscopies and office follow up visits, so healthcare utilization can be generally higher among these patients especially early in their disease course. Several internalized stigma subscales were associated with increased healthcare utilization. Specifically, the alienation subscale and the discrimination subscale were significantly associated with the number of outpatient visits and the number of endoscopies, respectively. Our group previously evaluated perceived stigma in EGIDs and found that increased overall perceived stigma was related to both increased outpatient visits and endoscopies (Guadagnoli et al., 2017). Further, patients with IBS experiencing internalized stigma, including the alienation subscale, also report increased healthcare utilization (Taft et al., 2014). One reason patients experiencing internalized stigma, specifically feelings of alienation, may seek additional healthcare is to feel a sense of validation. EGIDs are relatively new and rare (Dellon et al., 2014; Jensen et al., 2016), and have limited treatment options (Liacouras et al., 2011). Thus, it is not surprising that patients may change between multiple providers and seek additional or increased care in order to obtain adequate treatment. In a recent study evaluating unmet needs and barriers in adults and adult caregivers of children with EGIDs, Hiremath et al. (2018) found that a little over half of the respondents reported seeking multiple providers prior to their EGID diagnosis. Similarly, after being diagnosed with an EGID, 53% of respondents reported changing to a provider more knowledgeable about EGIDs (Hiremath et al., 2018). Further, in a qualitative study specifically assessing HRQOL in EG and EGE, a majority of the participants described seeking multiple providers prior to diagnosis and experiencing relief upon finding a provider knowledgeable about EG/EGE (Bedell et al., 2018). Increased healthcare utilization could also be attributed to ineffective treatments, which would require patients to continuously seek healthcare. In the current study, internalized stigma was associated with decreased treatment efficacy, indicating that EGIDs patients experiencing internalized stigma may perceive their treatments as ineffective, or ineffective treatments may encourage internalized stigma. In either case, patients experiencing stigma and decreased treatment efficacy may seek additional healthcare as a result.

Predictors of increased healthcare utilization were also assessed in the current study. When controlling for disease severity and symptom control, discrimination experiences predicted the number of endoscopies, while social withdrawal predicted the number of emergency department visits. EoE is the most common EGID (Dellon et al., 2014), resulting in EoE-majority samples in most studies including the present study. Endoscopies are regularly used to diagnosis and manage patients with EoE, as the major area affected in this subset of patients is the esophagus (Liacouras et al., 2011). In addition, some individuals with severe strictures may undergo esophageal dilation, an endoscopic procedure in which the esophagus is stretched out over multiple sessions. Therefore it is likely individuals with EoE are driving the increased endoscopies in the current study. However, individuals across the spectrum of EGIDs have reported experiencing food-related discrimination (Hiremath et al., 2018). It is possible that discriminatory experiences can further drive individuals to seek healthcare for frequent management and treatment of their EGID. On the other hand, it may be that the individuals requiring more endoscopic maintenance and treatment are reporting increased discrimination as a result. Regardless, endoscopies are financially taxing and inconvenient for EGID patients and caregivers (Hiremath et al., 2018). Thus, it is important for clinicians to consider the impact of stigma when working with patients, especially those requiring frequent endoscopies.

Social withdrawal was a significant predictor for number of ED visits. Previous studies have evaluated the negative impact of EGIDs on social functioning, including social withdrawal (Bedell et al., 2018). This may be due to feelings of embarrassment related to having a chronic condition or the symptoms associated with EGIDs (e.g. choking in public) (Taft, Kern, Kwiatek, et al., 2011). Individuals with EGIDs, in turn, may not tell partners, caregivers, or clinicians when they are experiencing symptoms in an attempt to conceal their disease or the severity of their symptoms. It is not uncommon for patients with a stigmatizing illness to attempt to “pass” as normal, and patients with EGID may engage in risky health behaviors (e.g. not adhering to their elimination diet) which then may result in ED visits when symptoms become intolerable. An additional way to conceptualize this would be that ED visits influence social withdrawal. It is possible that visits to the ED for an EGID, especially if they are frequent, can impact social functioning and lead to social withdrawal. Future research should further assess the directionality of the relationship between these variables, as understanding the mechanisms involved in the relationships between internalized stigma, increased healthcare utilization, and social impact may shed light on future opportunities for prevention and intervention.

Individuals on a dietary treatment at the time of the study reported increased internalized stigma in comparison to those not on a dietary treatment. Dietary modification is strongly recommended as a first-line treatment for EGIDs and includes the elimination and elemental diet (Dellon et al., 2013). The elimination diet requires individuals to remove common food triggers (e.g. milk, soy, egg, wheat, peanuts/tree nuts, shellfish/fish) from their diet for a period of time, then slowly reintroduce them as a way to identify potential triggers (Gonsalves et al., 2012). In elemental diets, individuals are limited to a diet of an elemental, or amino acid-based formula (Gonsalves, 2018; Liacouras et al., 2011). These diets can be costly and difficult to adhere to, and access to a registered dietician versed in EGID-specific dietary therapy may be a barrier to treatment for some (Hiremath et al., 2018). Adherence to the diet requires a change in food-related behaviors, such as planning meals ahead of time, eliminating favorite foods, and reviewing restaurant menus ahead of time (Bedell et al., 2018; Taft, Kern, Keefer, et al., 2011). Individuals may feel overwhelmed, anxious, or even anger and grief about these significant dietary changes (Bedell et al., 2018), which may in part explain increased stigma. However, those on dietary therapy at the time of the study did report increased HRQOL, which included the dietary and emotional subscales. Interestingly, past EoE literature indicates dietary modification is associated with reduced dietary-related QOL (Menard-Katcher et al., 2013; Stern, Taft, Zalewski, Gonsalves, & Hirano, 2018). One explanation for our findings could be that adherence to a diet reduces EGIDs symptoms, and therefore, patients are experiencing increased HRQOL. Research has found that adherence to a gluten-free diet in patients with celiac disease results in increased HRQOL, due to decreased symptoms and health concerns (Casellas et al., 2015; Casellas et al., 2008). Thus, it could be that our sample is experiencing symptom relief as a result of their diet, outweighing the cost of the restricted nature of the diet.

As expected, internalized stigma was associated with a decrease in HRQOL, and an increase in anxiety and depression. These findings are consistent with the internalized stigma literature in IBS and IBD (Taft et al., 2013; Taft & Keefer, 2016; Taft et al., 2014). In previous EGIDs literature, perceived stigma was also associated with decreased HRQOL and increased anxiety and depression (Guadagnoli et al., 2017). Further, it remained the strongest predictor of these three outcomes when controlling for gender, diagnosis duration, symptom control, and treatment efficacy (Guadagnoli et al., 2017). Subset analyses of the EoE-QOL-A found that internalized stigma was most strongly correlated with the emotional impact subscale. EGIDs alone can have a substantial impact on emotional functioning (Bedell et al., 2018; Hiremath et al., 2018; Taft, Kern, Keefer, et al., 2011). Thus, it is not surprising that incorporating stigma into one’s self-identity decreases emotional well-being.

Further demonstrating the negative impact of internalized stigma on HRQOL and emotional functioning is the finding that when controlling for disease severity and symptom control, alienation remained a significant predictor for HRQOL, anxiety, and depression. Alienation is comprised of three concepts: powerlessness (i.e. feeling helpless), normlessness (i.e. feeling out of place), and social isolation (i.e. feeling separated from the group) (Seeman, 1959). Individuals with EGIDs have endorsed experiencing social isolation, feeling like individuals without EGIDs do not understand their experience, and embarrassment or shame around their condition (Bedell et al., 2018; Hiremath et al., 2018; Taft, Kern, Keefer, et al., 2011; Taft, Kern, Kwiatek, et al., 2011). Further, the lack of public awareness, empathy, and supportive networks for their condition may contribute to feelings of alienation (Hiremath et al., 2018). Thus, it is important for providers to be aware of internalized stigma, particularly as it relates to alienation, and the influence it can have on patients psychosocial functioning.

Finally, individuals currently or previously engaged in therapy for issues related to their EGID report greater internalized stigma compared to those who have not sought EGID-related therapeutic treatment. Increased internalized stigma may drive individuals to pursue therapy as a way to cope with their disease and disease-related stressors. In Hiremath et al. (2018) 51% of respondents endorsed feeling the need to seek therapeutic help in managing their EGID. A promising finding from the current study is that those who have sought therapy for their EGID report increased emotional HRQOL, demonstrating the positive impact of psychological interventions can have on this patient population. Psychogastroenterology is a subset of health psychology specifically focused on the application of psychological interventions for symptom reduction and management in GI conditions (Keefer, Palsson, & Pandolfino, 2018). Incorporation of psychogastroenterology is imperative in delivering patient-centered gastroenterological care. These evidence-based interventions are effective in reducing symptoms and improving quality of life in a wide variety of GI illnesses, including IBS, IBD, functional heartburn, dysphagia, and globus sensation (Ballou & Keefer, 2017; Keefer et al., 2018; Riehl & Keefer, 2015; Riehl, Kinsinger, Kahrilas, Pandolfino, & Keefer, 2015). Given the clinical and psychosocial impact of EGIDs, including stigma, individuals with EGIDs may benefit from seeking a mental health professional with specific training in psychogastroenterology.

The present study has several limitations that should be taken into account when interpreting its results. First, we used one patient-reported question to capture each of the clinical variables of perceived disease severity, disease control, treatment efficacy, and treatment adherence. This is a limitation as the one question may not be an accurate reflection of the patient’s overall functioning. In addition, we did not control for comorbid factors, such as anxiety and depression, which could impact these clinical factors (e.g. treatment adherence). Future research should aim to include objective measures of clinical characteristics and/or measures that reflect patient behaviors. Our sample is a blend of clinic and online recruitment sources. As EGIDs are a rare condition, it was necessary to recruit from a diverse set of sources to ensure adequate statistical power. The use of online recruitment is growing in popularity and acceptability in both medical and psychological research (Lane, Armin, & Gordon, 2015; Topolovec-Vranic & Natarajan, 2016). However, we cannot confirm the EGID diagnosis in participants recruited online. It also may be likely that online recruits may feel more stigmatized about their disease (M. P. Jones, Bratten, & Keefer, 2007), thereby biasing our sample. The sample is also majority EoE, which reflects the known prevalence rates of EGIDs as a whole, so we may not have adequately captured the stigma experiences of EGID patients with non-EoE diagnoses. Finally, our sample is mostly White, female and college educated making the findings potentially difficult to apply to racial and ethnic minorities, and to men. EoE is more common in males, while EGE and ECE are more prevalent in females (Mansoor & Cooper, 2016; Mansoor, Saleh, & Cooper, 2017). The lack of male representation in our sample demonstrates a limitation, especially given that EoE is the most common EGID. Further, although EGIDs are primarily seen in White individuals, research suggests there may be differences in symptoms and physiological features by race, underscoring the importance of diversification in EGID research (Mansoor et al., 2017; Moawad et al., 2016). Future studies should seek to recruit more diverse samples across these limitations to better understand how internalized stigma may present in EGIDs.

This is the first study to evaluate internalized stigma in EGIDs, a relatively rare but growing patient population in gastroenterology practice. Like other diseases, internalized stigma has the potential to increase psychological distress, reduce quality of life, and increase healthcare utilization and cost. Clinicians working in gastroenterology should assess feelings of stigma, such as social withdrawal and alienation, in their EGID patients. Referral for psychogastroenterology and other GI-related behavioral medicine services can help mitigate some of the negative impacts of internalized stigma via evidence-based psychological interventions.