Disordered eating and eating competence in members of online irritable bowel syndrome support groups.

This cross-sectional study is based on an anonymous online survey conducted from August to September 2021. Individuals were eligible to participate if they were aged 18 or older, could read and understand English, had not taken the survey before, and self-reported being diagnosed with IBS by a physician. After pilot-testing for functionality and face validity, the survey link was posted to the following online and social media forums: “IBS Support (Official)” on Facebook.com , Reddit.com/r/ibs , Ibsgroup.org , and Ibspatient.org . The survey was accessed on REDCap, a secure web-based platform hosted through the University’s Institute of Translational Health Sciences. 27 , 28 Exempt status was granted by the University’s Human Subjects Division. The first part of the survey collected information about participants’ age, height, weight, gender, sex assigned at birth, sexual orientation, race, country of residence, IBS subtype, years since IBS diagnosis, ED history, and comorbidities. The validated IBS Severity Scoring System (IBS-SSS) was used to assess IBS symptom severity. 29 The IBS-SSS contains five questions that pertain to abdominal pain, abdominal distention, dissatisfaction with bowel functioning, and interference of IBS symptoms with life. Response options are provided in increments of 10. Scores 75–174 indicate “mild IBS”, scores 175– 299 indicate “moderate IBS”, and scores ≥300 indicate “severe IBS”. Individuals who self-reported a diagnosis of IBS but scored <75 were included in the mild group. The validated ecSI 2.0 ™ was used to assess EC. 20 Individuals scoring ≥32 on ecSI 2.0 ™ are classified as eating competent. 17 ecSI 2.0 ™ has four subscales or domains: eating attitudes, internal regulation, food acceptance, and contextual skills. 20 All items are scored on a 5-point Likert scale with the following choices and point values: “Always” (3), “Often” (2), “Sometimes” (1), “Rarely” (0), and “Never” (0). Items are summed to determine ecSI 2.0 ™ scores. The validated EAT-26 was used to assess disordered eating likelihood and has three subscales: dieting, bulimia and food preoccupation, and oral control. 30 , 31 Items 1–24 and 26 are scored on a 6-point Likert scale with the following choices and point values: “Always” (3), “Usually” (2), “Often” (1), “Sometimes” (0), “Rarely” (0), and “Never” (0). Item 25 is reverse scored. Items are summed to determine EAT-26 score. Scores 0–10 correspond to “Normal; modest to low anxiety,” scores 11–20 correspond to “Moderate or ‘normative’ disturbance,” scores 21–30 correspond to “Significant disturbance, likely disordered eating,” and scores ≥31 correspond to “Extreme disturbance, very likely disordered eating.” One optional open-ended question was included at the end of the survey, “Is there anything else you would like to share with the research team?” Preliminary analyses were conducted to determine whether gender and age were relevant covariates for the models. Two-way ANOVAs with Bonferroni post hoc tests for multiple comparisons were used to examine the relationships between IBS subtype, IBS severity level, and ecSI 2.0 ™ total and subscale scores. The first grouping variable was IBS severity with three levels: mild (n=34), moderate (n=82), and severe (n=90). The second grouping variable was IBS subtype with three levels: IBS with constipation (IBS-C) (n=31), IBS with diarrhea (IBS-D) (n=74), and IBS with alternating or unclearly reported bowel habits (Unknown) (n=101). The interaction between IBS severity level and IBS subtype in relation to ecSI 2.0 ™ total score was explored. In instances where ANOVA assumptions were not fully met, the non-parametric Kruskal-Wallis H test was used to verify results. No discrepancies between parametric and non-parametric results were noted, so parametric results are reported. Multiple linear regression was used to predict EAT-26 total and subscale scores from gender (if applicable), IBS-SSS score, and age. Each of the four regression models were then repeated with IBS subtype added as an additional covariate. The assumptions of normality, linearity, independence of residuals, homoscedasticity, and no multicollinearity were met for all models. SPSS version 28.0 was used for all statistical analyses. 32 The significance level was 0.05.

260 individuals followed the survey link, 225 of whom met the eligibility criteria and indicated their informed consent by proceeding to the questionnaires. Table 1 displays a summary of participant demographics and relevant health information. Table 2 displays a summary of participant IBS-SSS, ecSI 2.0 ™ , and EAT-26 scores and categorizations. One-way ANOVAs and t -tests revealed that age was negatively associated with EAT-26 total ( p <0.001) and subscale scores (all p <0.05) and that gender was related to EAT-26 oral control subscale score ( p =0.007). Thus, age was incorporated into all EAT-26 models and gender was incorporated into the EAT-26 oral control subscale model. Gender and age were not significantly related to ecSI 2.0 ™ total or subscale scores (all p >0.05) and thus were not included in the ecSI 2.0 ™ models. There was a main effect of IBS severity level on ecSI 2.0 ™ score ( p =0.022), but no relationship between IBS subtype and ecSI 2.0 ™ score ( p =0.938). As shown in Figure 1 , the mean (±SD) ecSI 2.0 ™ score for the severe group (21.1±9.3) was significantly lower ( p =0.016) than the moderate group (24.9±9.1) but not significantly different from the mild group (24.1±9.2). There was no interaction between IBS severity level and IBS subtype in relation to ecSI 2.0 ™ total score ( p =0.596). There was a main effect of IBS severity level on ecSI 2.0 ™ eating attitudes subscale score (p=0.003), but no relationship with IBS subtype (p=0.599). There were no main effects for any of the other subscales (all p>0.05). There was no interaction between IBS severity level and IBS subtype in relation to each of the ecSI 2.0 ™ subscales, all p >0.05. IBS-SSS positively associated with EAT-26 total score ( p =0.011) and EAT-26 subscale scores on dieting ( p =0.001) and bulimia and food preoccupation ( p =0.002). Age inversely associated with EAT-26 total score ( p 0.05) when the models included IBS subtype as an additional covariate. Regression coefficients and standard errors are in Table 3 .

This study found a positive association between IBS severity and likelihood of disordered eating, and EC was significantly lower in the severe IBS group compared to the moderate IBS group. No relationship was detected between IBS subtype and likelihood of disordered eating or EC. Only 17.1% of the sample was classified as eating competent, 27.0% was classified as likely or very likely disordered eating, and 29.1% self-reported having a past or present ED. The participants were predominantly females with moderate or severe IBS. The predominance of females in the sample aligns with existing research, which shows that IBS rates in women are 1.5- to 3-times the rate of IBS in men. 33 Further, there may be sex differences in willingness to seek support on social media. This study adds to the body of research exploring the relationship between IBS severity and likelihood of disordered eating. The inverse relationship between EAT-26 score and age and the positive relationship between EAT-26 score and IBS severity aligns with work done by Kayar et al. 4 Other studies have provided conflicting evidence. One found no association between SCOFF score and IBS-SSS. 2 Another found that IBS severity was not associated with total EDI score, though it was associated with multiple EDI subscales. 8 These differing findings may be due to heterogeneity of assessment tools. More research is needed to develop a clear understanding of the role IBS symptom severity plays in the likelihood of disordered eating. The rate of EC observed in this sample (17%) is lower than rates of EC in other populations. For comparison, the rate of EC was 45% in a sample of low-income females 24 , 47% in a sample of college students 17 , and 53% in a sample of parents of 4 th graders. 34 This indicates that a greater proportion of patients with IBS struggle to maintain a flexible and positive approach to eating than in more general populations. While EC has not previously been assessed in a sample of patients with IBS, the low rate of EC in this sample aligns with existing research on the impacts of IBS on eating behaviors. Irregular eating habits such as skipping meals and going extended periods of time without eating have been noted in individuals with IBS, particularly females. 22 , 23 In one study, 42% of adolescents surveyed reported “not eating any food even when hungry” to control symptoms. 7 These findings align with the contextual skills and internal regulation subscales of ecSI 2.0 ™ , as well as with several open-ended survey responses in our study. One participant stated, “I tend to starve myself to avoid triggering my IBS” and another reported that they “only eat one meal per day to keep from living in the bathroom.” These comments illustrate how IBS can influence eating patterns and attunement to appetite cues, impacting overall EC. The fact that the severe IBS group had the lowest EC aligns with existing evidence related to dietary triggers and food avoidance in patients with IBS. A recent study found a positive association between IBS severity and the number of foods patients report as triggers for their IBS. 35 Another study found that severe food avoidance and restriction in IBS patients was associated with greater symptom severity, lower quality of life, and lower energy and nutrient intake. 3 Food acceptance in this population may be impacted by the use of elimination diets, a common nutrition intervention for IBS that may aid in symptom control 36 , or by fear and anxiety surrounding food choices. One study participant indicated in their open-ended response that they are “terrified of trying new foods,” illustrating the role that fear can play in lower food acceptance. While previous research has not directly examined enjoyment of food in individuals with IBS, our study demonstrates an inverse relationship between IBS symptom severity and the ecSI 2.0 ™ eating attitudes subscale, which encompasses overall enjoyment of food. Several responses from the open-ended survey question illustrate lower enjoyment of food, including one participant who stated, “The pleasure on my tastebuds is ruined by the fear of what the food will do in my intestines.” The effects of IBS on eating attitudes and enjoyment may impact patients’ quality of life and is an area for further research. Strengths of this study include the use of validated questionnaires and homogenous recruitment methods. Limitations must be considered. There is a lack of ethnic and racial diversity in this sample, with 84.8% of participants reporting their race/origin as White or Caucasian. Participants self-reported a diagnosis of IBS and in some cases reported comorbidities including other GI disorders. It is possible that these comorbidities impacted disordered eating likelihood and eating competence. Further, the sample is limited to individuals who have viewed or joined an online IBS support forum. These individuals may have shared characteristics that differ from the general population of patients with IBS, such as heightened concerns about their health or greater symptom severity. This may have contributed to the small number of participants in the mild IBS group, which may have been underpowered to make accurate comparisons to the moderate IBS and severe IBS groups. Additionally, individuals who self-selected to participate in the survey may have unique characteristics, such as access to the internet and an interest in participating in research. Because this study is cross-sectional, directionality and causality of relationships cannot be established. From a statistical perspective this is an exploratory study. Therefore, no allowance has been made for the number of statistical hypothesis tests reported, and any p-value <0.05 has been taken as evidence to reject the null hypothesis. Since this approach may result in false rejection of some null hypotheses, the findings of this study need to be replicated to be considered statistically robust. The results of this exploratory study highlight the importance of awareness of disordered eating and EC in clinical gastroenterology practice and the need for further studies in this area. Given that highly restrictive diets are contraindicated in patients with an active ED, 37 the knowledge that the likelihood of disordered eating may be higher with higher IBS symptom severity should be considered when selecting therapies. The presence of “likely” or “very likely” disordered eating in over one-fourth of our sample sheds light on the importance of screening for EDs in this population so that appropriate referrals for ED treatment can be made when necessary, and so that support can be provided to prevent those in high-risk categories from progressing to a clinical ED. This study also calls attention to the relatively low rates of EC in patients with IBS, particularly among females with severe IBS. Intervention studies can help determine if improving EC in patients with IBS improves symptom control. It is possible that focusing on factors such as meal timing and honoring hunger, appetite, and satiety cues may promote digestive regularity and aid in normalizing colonic motility. 22 Promoting relaxed and positive attitudes about food could potentially lower stress and anxiety at mealtimes and thereby modulate the gut-brain axis. 38 Eliminating overly restrictive eating patterns through a focus on EC principles may improve overall nutrition status 24 and decrease stress 26 and binge eating behaviors, 39 both of which may exacerbate IBS symptoms. 38 , 40

Irritable bowel syndrome (IBS) is one of the most common disorders of gut-brain interaction (DGBI), affecting an estimated 11% of the global population. 1 It is a chronic condition characterized by abdominal pain and altered bowel habits that are associated with significant reductions in quality of life. 1 Existing evidence suggests that disordered eating likelihood is high among patients with IBS and other DGBIs. In cross-sectional studies using a variety of validated tools including the 26 Item Eating Attitudes Test (EAT-26), the 40 Item Eating Attitudes Test (EAT-40), the Eating Disorder Inventory (EDI), the Eating Disorders Examination Questionnaire (EDE-Q), and the SCOFF Questionnaire, 23–48% of patients with IBS screened positive for disturbed eating patterns or possible eating disorder (ED) symptoms. 2 – 8 These symptoms include self-induced vomiting, weight and shape concerns, drive for thinness, dietary restraint, food preoccupation, and emotional dysregulation. 2 – 8 Studies examining the broader population of patients with DGBIs found that approximately 20% screened positive for possible avoidant/restrictive food intake disorder (ARFID) based on their scores on the Nine Item ARFID Screen (NIAS). 9 , 10 ARFID is characterized by persistent failure to meet caloric needs in the absence of fear of weight gain. Patients with ARFID avoid eating for other reasons, such as fear of symptoms after eating. 10 Few studies have measured the likelihood of disordered eating in IBS patients compared to controls. Two studies found that patients with IBS scored significantly higher on EAT-40 compared to patients with inflammatory bowel disease or peptic ulceration and healthy controls, indicating a higher level of concern about dieting and body weight among the IBS groups. 5 , 6 Other studies found no significant differences in EDE-Q and EDI scores between patients with IBS and healthy controls, though a higher rate of specific disordered eating behaviors (e.g., skipping meals and vomiting after eating as a means of controlling symptoms) and thoughts of vomiting as a means of weight control were noted. 7 , 8 EAT-26 scores have been shown to be higher among patients with IBS compared to controls, and IBS severity was positively associated with EAT-26 scores. 4 Overall, existing evidence suggests that patients with IBS exhibit more warning signs for disordered eating compared to patients with organic gastrointestinal (GI) disorders and healthy controls. The relationship between EDs and GI disorders may be bidirectional in nature. Avoidance of certain foods is reported in up to 98% of patients with IBS. 7 Research indicates that 15.9% of patients with DGBI have a history of ED and 87% of patients with ED and IBS developed their ED prior to IBS, indicating that in some cases ED precedes DGBI diagnosis. 11 , 12 Disordered eating practices such as self-induced vomiting, binge eating, laxative abuse, and caloric restriction have the potential to lead to physiological changes including esophageal acid damage, changes in hormone secretion, damage to pelvic floor muscles, delayed gastric emptying, and abnormal colonic function that may result in GI symptoms. 13 However, there is a positive association between GI problems in childhood and early adolescence and later disordered eating, suggesting that in other cases, DGBI precede ED diagnosis. 14 , 15 This may be due to restrictive dietary changes that have the potential to progress to a maladaptive focus on food intake and ultimately disordered eating, as proposed in Satherley et al.’s conceptual model of disordered eating in GI disorders. 16 In contrast to disordered eating, eating competence (EC) describes a flexible, positive approach to consuming nutritious and enjoyable food. 17 ED symptoms are inversely related to scores on the Satter Eating Competence Inventory (ecSI 2.0 ™ ) 17 – 19 , a validated questionnaire that measures EC. 20 No existing research assesses ecSI 2.0 ™ in patients with IBS. While other models that encourage attunement to appetite have been more frequently studied in relation to EDs, such as intuitive eating and mindful eating, EC differs in that it also encompasses the contextual skills related to planning for, preparing, and consuming meals and snacks on a regular schedule. 21 The contextual skills domain of the EC model may be particularly relevant in IBS due to the fact that patients with IBS report skipping meals and going long periods of time without eating at a higher rate than healthy controls. 22 , 23 There may also be a greater need for contextual skills in patients with dietary restrictions related to IBS to ensure nutritional adequacy. Further, given that EC is associated with a higher Healthy Eating Index, improved metabolic health markers, lower stress, and lower body dissatisfaction 24 – 26 , it would be valuable to study EC in individuals with IBS. This exploratory study seeks to add new knowledge on the likelihood of disordered eating in members of online IBS support groups, evaluate their level of EC, and determine whether EC and likelihood of disordered eating varies according to IBS symptom severity and subtype; the interaction between these two factors will be explored. We predict an inverse relationship between IBS severity and EC, and a positive association between IBS severity and disordered eating likelihood.