Binge eating disorder (BED) is a syndrome that falls within the spectrum of eating disorders with diagnostic criteria established by the Diagnostic Statistical Manual of Mental Disorders-5 (DSM-5). The diagnostic criteria define it as the presence of binge eating episodes that occur at least once a week for the past 3 months prior to evaluation and are not associated with the diagnosis of bulimia nervosa, anorexia nervosa, or inappropriate compensatory behaviors (eg, purging, fasting, or excessive physical activity).¹ The worldwide prevalence of eating disorders is on the rise, with an estimate of 2% lifetime prevalence.^2-4 Eating disorders (EDs) are associated with the presence of functional dyspepsia (FD), showing a higher prevalence of this disorder of gut-brain interaction (DGBIs) compared to the general population, in particular its syndromic variety of postprandial distress syndrome (PDS).^5-8 It has been documented that some of the most prevalent dyspeptic symptoms referred to by patients that present binge-eating behavior are postprandial fullness, epigastric pressure, and nausea.⁵ There is evidence supporting the hypothesis of a pathophysiological association between BED and gastroduodenal DGBIs, with demonstrated variability in gastric volume capacity, as well as dysregulation of hunger and satiety in BED patients.^7,8

It has been confirmed that many patients with EDs will exhibit a DGBI, which may persist even after remission of the initial ED. This suggests the existence of a potential vicious cycle, wherein dyspeptic symptoms such as bloating and early satiation hinder patients from normalizing their eating behaviors and may even perpetuate disordered eating.^6,7,9,10 The existence of dyspeptic symptoms in individuals with ED significantly influences their quality of life, implying it as a therapeutic focus corresponding to the prevailing syndrome. Despite evidence of this correlation between these 2 conditions, scant information exists regarding BED, owing to its recent delineation and revision in diagnostic criteria stipulated by the DSM-5 in contrast to its prior depiction in DSM-4.

Patients

The study was conducted in 7 states of Mexico under the coordination of gastroenterologists, with a defined timeframe from September 1st, 2021, to November 14, 2022. The implementation of the protocol adheres to the recommendations outlined in the Universal Declaration on Bioethics and Human Rights, adopted on October 19, 2005, during the UNESCO General Conference. This declaration stipulates that no known or unforeseen risks should be posed to the subjects participating in the investigation protocol, ensuring the confidentiality of participants and the information obtained. The study is an analysis of the database from the protocol registered with the local ethics committee under folio 113/21. Informed consent was obtained verbally by the surveying physicians and written informed consent was provided on paper.

The study included individuals over 18 years old, excluding those with organic pathologies that could account for gastrointestinal symptoms. Subjects who met the criteria for BED according to DSM-5 recommendations were selected for comparative analysis. Patients suspected of having anorexia or bulimia were excluded based on the definition of BED.

Data Collection and Variables

The collection of data was performed through the application of an electronic survey applied online to an open population. The patients were collected by probability sampling with a cluster design. Sociodemographic data, including medical history of comorbidities and current medical management, gastrointestinal symptomatology related to esophageal, gastroduodenal, bowel, and anorectal disorders according to the Rome IV recommendations, including in these the diagnostic criteria of FD and its subtypes, as well as information required for BED diagnosis according to DSM-5.¹¹

For the assessment of the gastrointestinal symptomatology severity a 7-point Likert scale was applied for each symptom, considering a score of > 6 points as severe symptomatology. The screening for anxiety and depression was conducted using the Hospital Anxiety and Depression Scale in its validated Spanish version in a Mexican population. Scores of ≥ 11 points in the sub-analysis for both pathologies were considered indicative of anxiety or depression.¹²

Statistical Methods

An analytical cross-sectional study was performed applying a comparison between 2 groups, one comprising patients that met BED criteria and the other consisting of patients without BED diagnosis. For the analysis of independent nominal and ordinal qualitative variables, a multivariate logistic regression analysis with the χ² test was conducted, and results were reported as odds ratios (OR) with 95% confidence intervals (CI), considering a statistically significant result with P-values < 0.05. Continuous variables were subjected to distribution analysis using the Kolmogorov-Smirnov and Shapiro-Wilk tests, as well as an equality of variances analysis using the Levene test. The recommended test based on the distribution and homoscedasticity of the variables was used for comparison between both groups. For quantitative variables with normal distribution and homoscedasticity, the t Student test or ANOVA was employed, while Welch’s correction was used in cases of heteroscedasticity. For quantitative variables with non-normal distribution and homoscedasticity, the Mann-Whitney U test or Kruskal-Wallis test was applied, with variable transformation conducted in cases of heteroscedasticity. Continuous variables were expressed as means with standard deviation or median with interquartile range. Statistical analysis was performed using IBM SPSS Statistics version 28.0.1 software and R Project for Statistical Computing software (R Foundation for Statistical Computing, Vienna, Austria).

Demographic

A total of 1016 subjects with a mean age of 27 ± 9 years and a female predominance of 68.8% were included in the study, of whom 373 (36.7%) met the criteria for BED. We observed a significant difference in age (BED: 26 ± 7 vs non-BED: 28 ± 9 years, P < 0.001), but not in gender distribution (BED: 66% female vs non-BED: 70% female, P = 0.177). When comparing body mass index (BMI), both groups tended towards overweight; however, BED patients exhibited a slightly lower BMI (26.6 ± 4.3 vs 27.0 ± 3.7 kg/m², P = 0.001) (Table 1).

Table 1 . Demographics and Clinical Characteristics

Variable	BED (n = 373)	Non-BED (n = 643)	P-value
Age (yr) 18-34 35-49 50-64	26.4 ± 7.5 334 (89.5) 31 (8.3) 8 (2.1)	28.0 ± 9.9 528 (82.1) 69 (10.7) 46 (7.2)	< 0.001 0.001 0.212 0.001
Sex
Female Male	247 (66.2) 137 (36.7)	452 (70.3) 191 (29.7)	0.177 0.177
WHO Classification of body mass index
Normal range Overweight Obesity grade I Obesity grade II Body mass index (kg/m2)	137 (36.7) 129 (34.6) 100 (26.8) 7 (1.9) 26.6 ± 4.3	221 (34.4) 248 (38.6) 168 (26.1) 6 (0.9) 27.0 ± 3.7	0.448 0.205 0.812 0.197 < 0.001
Dyspepsia symptoms
Epigastric pain Epigastric burning Early satiation Postprandial fullness	135 (36.2) 112 (30.0) 144 (38.6) 136 (36.5)	169 (26.3) 138 (21.5) 245 (38.1) 148 (23.0)	0.001 0.002 0.874 < 0.001
Dyspepsia Epigastric pain syndrome Postprandial distress syndrome Overlap syndrome	200 (53.6) 44 (11.8) 65 (17.4) 91 (24.4)	282 (43.9) 71 (11.0) 112 (17.4) 99 (15.4)	0.003 0.715 0.997 < 0.001
Dyspeptic symptom severity
Epigastric pain syndrome 7-point Likert scale Postprandial distress syndrome 7-point Likert scale	2.23 ± 0.16 2.70 ± 2.28	1.7 ± 0.12 2.28 ± 1.80	0.018 0.027
Hospital Anxiety and Depression Scale
Anxiety Depression	181 (48.5) 100 (26.8)	209 (32.5) 113 (17.6)	< 0.001 < 0.001

BED, binge eating disorder; WHO, World Health Organization.

Data are presented as mean ± SD or n (%).

Clinical Features

The prevalence of dyspepsia in BED was 53.6% (95% CI, 48.4-58.7) (Fig. 1). Epigastric pain syndrome (EPS) was present in 11.8% (95% CI, 8.7-15.5), PDS in 17.4% (95% CI, 13.7-21.6), and 24.4% (95% CI, 20.1-29.0) had overlap (OVL) (Fig. 2). Epigastric burning regularly affected 112 (30%) of BED subjects while epigastric pain affected a higher proportion, with 135 (36.2%) subjects experiencing this symptom. Among the symptoms of PDS, early satiety was the most common, affecting 144 (38.6%) subjects, followed by postprandial fullness, which affected 135 (36.2%) subjects (Table 1).

Figure 1. Prevalence of functional dyspepsia in a Mexican population with binge eating disorder.

Figure 2. Prevalence of functional dyspepsia subtypes in a Mexican population with binge eating disorder.

On the contrary, individuals without BED showed dyspepsia in a lower proportion, with a prevalence of 43.8% (95% CI, 39.9-47.7) (Fig. 1). The prevalence of EPS and PDS was similar to that calculated in patients with BED, being 11.0% (95% CI, 87.3-13.7) and 17.4% (95% CI, 14.5-20.5) respectively; however, the prevalence of OVL was lower in subjects without BED, being 15.4% (95% CI, 12.6-18.4) (Fig. 2).

The diagnosis of anxiety affected nearly half of the patients with BED, being present in 181 (48.5%) of these individuals. In contrast, the diagnosis of depression was only present in 100 (26.8%) of the patients with this eating behavior disorder (Table 1).

Comparing Patients With and Without Binge Eating Disorder

In a multivariate logistic regression analysis, 18-34 years age group (OR, 3.01; 95% CI, 1.38-6.58; P = 0.005) and male sex was found to be independently associated with BED (OR, 1.40; 95% CI, 1.05-1.88; P = 0.021). There was no association found between BED and WHO Classification of BMI (Table 2).

Table 2 . Comparative Multivariable Analysis: Binge Eating Disorder Versus Non-binge Eating Disorder

Variable	OR	95% CI	P-value
Age (yr)
18-34 35-49 50-64	3.019 2.214 0.343	1.385-6.583 0.917-5.348 0.158-0.747	0.005 0.077 0.007
Sex
Female Male	0.70 1.40	0.52-0.94 1.05-1.88	0.020 0.021
WHO Classification of body mass index
Normal range Overweight Obesity grade I Obesity grade II	0.97 0.66 0.58 1.59	0.29-3.22 0.20-2.18 0.17-1.92 0.49-5.15	0.972 0.504 0.380 0.432
Dyspepsia symptoms
Epigastric pain Epigastric burning Early satiation Postprandial fullness	1.33 1.49 0.70 1.52	0.87-2.03 0.74-0.74 0.49-0.99 1.06-2.17	0.186 0.263 0.045 0.021
Dyspepsia Epigastric pain syndrome Postprandial distress syndrome Overlap syndrome	1.29 1.29 0.91 1.80	0.98-1.71 0.82-2.02 0.61-1.35 1.25-2.60	0.066 0.268 0.640 0.002
Hospital Anxiety and Depression Scale
Anxiety Depression	1.74 1.30	1.26-2.39 0.89-1.88	0.001 0.165

Among the dyspeptic symptoms postprandial fullness (OR, 1.52; 95% CI, 1.06-2.17; P = 0.021) was associated with BED. When analyzing the comparison of dyspeptic syndromes, only OVL was associated with BED (OR, 1.80; 95% CI, 1.25-2.60; P = 0.002). Regarding psychiatric pathology, only anxiety was statistically significantly associated with BED (OR, 1.74; 95% CI, 1.26-2.39; P = 0.001) (Table 2).

When comparing symptom severity using a 7-point Likert scale, significant difference was found in the severity of EPS (2.23 ± 0.16 vs 1.7 ± 0.12 points, P = 0.018) and PDS symptoms (2.70 ± 2.28 vs 2.28 ± 1.80 points, P = 0.027), presenting patients with BED exhibiting more severe symptoms (Table 1).

Comparing Binge Eating Disorder Patients With and Without Dyspepsia Overlap

When comparing patients with an overlap of BED and dyspepsia with those who solely presented with BED, there was no association between group of age, sex or World Health Organization Classification of BMI. An association was found between the group of BED-dyspepsia overlap and depression (OR, 2.03; 95% CI, 1.12-3.67; P = 0.019) (Table 3).

Table 3 . Comparative Multivariable Analysis: Binge Eating Disorder With Dyspepsia Versus Binge Eating Disorder Without Dyspepsia

Variable	OR	95% CI	P-value
Age (yr)
18-34 35-49 50-64	2.65 7.18 0.34	0.49-14.3 1.11-46.4 0.06-1.83	0.257 0.038 0.211
Sex
Female Male	2.60 1.40	1.62-4.17 1.05-1.88	< 0.001 0.021
WHO classification of body mass index
Normal range Overweight Obesity grade I Obesity grade II	0.24 0.40 0.23 3.92	0.26-2.27 0.04-3.75 0.26-2.21 0.42-36.20	0.215 0.426 0.207 0.229
Hospital Anxiety and Depression Scale
Anxiety Depression	1.56 2.03	0.92-2.65 1.12-3.67	0.093 0.019

The prevalence of DGBIs in populations diagnosed with EDs reaches up to 98%, with FD being one of the most common, as reported by Santonicola et al,⁵ who found a prevalence of PDS in EDs of 90%.^13,14 Despite this, most literature reports an association of BED with other EDs such as anorexia nervosa and bulimia nervosa, but there is limited information regarding the prevalence and association between dyspepsia and BED.^5,13 In a study by Cremonini et al,¹⁴ a BED prevalence of 6.1% was reported, contrasting dramatically with our findings, where a high prevalence of this disorder was found (36.7%; 95% CI, 33.7-39.7). However, different diagnostic methods were used; in their study, the Questionnaire on Eating and Weight Patterns was used, while in ours, the criteria established by the DSM-5 were employed. These criteria have reported a BED prevalence of 23.0% in healthy populations and up to 30.5% in populations with comorbidities that impact satiety regulation, such as polycystic ovary syndrome.

It is noteworthy that there are no studies on BED prevalence in dyspepsia; however, due to its association with anxiety and its impact on hunger and satiety regulation, a hypothesis of a higher prevalence of BED than in the general population was considered.^14-17

They also reported an association between BED and upper abdominal pain, a finding not observed in our study, which instead found an association with postprandial fullness (OR, 1.52; 95% CI, 1.06-2.17; P = 0.021). However, in syndromic assessment, an association with OVL was found (OR, 1.80; 95% CI, 1.25-2.60; P = 0.002).¹⁴ According to the meta-analysis published by Barberio et al,¹⁸ the overall prevalence of dyspepsia is 6.9% (95% CI, 5.7-8.2), like a multicenter study in the Mexican population published by Remes-Troche et al,¹⁹ where a prevalence of 12% was found.^17,18 When comparing the results, the higher prevalence of dyspepsia in patients with BED is evident, with our study finding a prevalence of 53.6% (95% CI, 48.4-58.7).^18,19

Despite females predominating in both groups of the studied population, when comparing the frequency of females in both groups, it was similar (BED: female 66.2% vs Non-BED: female 70.3%). However, the difference in the frequency of males between the 2 groups was greater (BED: male 36.7% vs Non-BED: male 29.7%) (Table 1). Although these differences were not significant in a univariable analysis, a multivariable logistic regression analysis revealed an association between BED and male sex (OR, 1.40; 95% CI, 1.05-1.88; P = 0.021) (Table 2).

It is remarkable that an association was found between BED and the PDS symptoms, early satiety and postprandial fullness; however, no association was found with PDS. This finding can be explained by a significant proportion of patients with early satiety and postprandial fullness being classified within the OVL group rather than the pure PDS group (Table 1).

It can be concluded that the prevalence of BED is higher in the Mexican population, with a greater prevalence of dyspepsia found in this population. BED is associated with symptoms of PDS such as postprandial fullness; however, the predominant syndromic phenotype is OVL. Patients presenting this OVL are predominantly female with depression.

Hypotheses suggest that the correlation between EDs and DGBIs stems from alterations in gastric emptying and accommodation⁷. By demonstrating that patients with BED exhibit a larger gastric capacity compared to non-BED patients, along with alterations in hunger and fullness perception, it can be inferred that these factors exacerbate binge episodes. This, in turn, may lead to impaired gastric emptying and accommodation, further reinforcing the idea that disordered eating significantly contributes to the development of dyspeptic symptoms.⁸ Another proposed mechanism involves the secretion of dopamine triggered by the consumption of highly processed foods, which mirrors the reward responses seen in addiction. Additionally, imaging studies have indicated an increase in orbitofrontal cortex volumes in patients with BED compared to healthy individuals which may be associated with dysfunctions in food reward processing and self-regulation.^20,21

Despite binge eating episodes in BED potentially explaining symptoms of PDS, it has also been demonstrated that the symptomatic severity of ED correlates with the symptomatic severity of coexisting DGBIs and anxiety.²² Supporting this hypothesis, there is evidence that patients with FD exhibit altered eating behaviors without meeting criteria for any EDs, as shown in a study by Shimpuku et al,²³ where patients with FD demonstrated more frequent between-meal eating episodes compared to healthy controls. Given this, it could be inferred that the presence of FD may also be a risk factor for developing BED; however, prospective studies are needed to determine this association.

Limitations

The diagnosis of FD was based on the patient^’s symptoms, applying the Rome IV criteria; however, no studies were conducted to rule out organic pathology that could explain these symptoms. Nevertheless, individuals with personal pathological backgrounds that could explain such symptomatology were excluded from the study.

As this is a cross-sectional study, only the prevalence of dyspepsia in BED can be determined. However, establishing a causal factor is not possible since it is not objectively feasible to ascertain which of the 2 disorders manifested first. Prospective studies of dyspepsia and BED would be necessary to establish an association measure that considers temporality.

Given that the prevalence of EDs in patients with DGBIs is higher than in the general population, clinical suspicion and screening for EDs in these patients should be part of the initial medical assessment. Therefore, considering that the symptomatic severity of other EDs such as Avoidant/Restrictive Food Intake Disorder may influence the development or symptomatic severity of DGBIs like FD, not applying diagnostic criteria for other EDs is a limitation of our study. However, the primary aim was to determine the association between BED and FD, and administering additional questionnaires in an online survey could have been burdensome and impractical for the surveyed population.

We would like to express our sincere gratitude to the entire research team involved in this study for their dedication, hard work, and valuable contributions. We are also deeply appreciative of the patients who participated in this study. Their willingness to contribute their time and experiences was essential to the success of our research.

None.

None.

Francisco A Félix Téllez, Alejandra X Cruz Salgado, and José A Velarde Ruiz Velasco were responsible for the organization, planning, and logistics of the research; Ángel R Flores Rendon, Marco A O Flores Lizárraga, José I Soto González, and Nadia S Abizaid Herrera were in charge of survey administration; Héctor R Ordaz Álvarez and José M Remes Troche handled the statistical analysis; and Francisco A Félix Téllez took charge of manuscript writing.