Abstract
Background
Individuals with inflammatory bowel disease (IBD) often modify their diet to manage symptoms; however, these behaviors may evolve into eating disorders, including avoidant/restrictive food intake disorder (ARFID). We assessed the risk of eating disorders and ARFID in patients with IBD compared with healthy controls (HC), explored differences by age at diagnosis, and examined associations with malnutrition and disability.
Methods
In this cross-sectional study, adult patients with confirmed IBD, stratified by pediatric- vs adult-onset, were matched with HC. ARFID risk was assessed using the Nine-Item ARFID Screen (NIAS-9) and eating disorders risk with the Eating Attitudes Test-26 (EAT-26). Nutritional status was evaluated with the Patient-Generated Subjective Global Assessment (PG-SGA) and disability with the IBD-Disk and modified IBD-Disk.
Results
A total of 706 participants completed questionnaires (355 IBD, 351 HC). Eating disorder risk did not differ between groups (11.3% vs 10.8%, P = .91). ARFID risk was higher in IBD (13.5% vs 5.7%, P < .001), with more fear-driven eating (P < .001) and lower picky eating (P < .001) and appetite scores (P = .033). ARFID risk did not differ by age at onset (P = .39). Independent associated factors included active disease (odds ratio [OR] 2.34, 95% confidence interval [CI] 1.10-5.01), malnutrition (OR 2.31, 95% CI 1.04-5.13), dietary changes (OR 4.32, 95% CI 1.92-9.74), and eating disorder risk (OR 7.47, 95% CI 2.95-18.90). Even in remission, ARFID risk remained elevated compared to HC (10.1% vs 5.7%, P = .03).
Conclusions
ARFID risk in IBD is nearly twice that of HC and strongly associated with disease activity, malnutrition, and disability, supporting the importance of ARFID screening in routine IBD care.
Keywords: ARFID, food avoidance, eating disorders, malnutrition, disability, inflammatory bowel disease
Graphical abstract
Graphical Abstract.
1. Introduction
Inflammatory bowel diseases (IBD) represent a group of chronic, relapsing–remitting conditions that include Crohn’s disease (CD) and ulcerative colitis (UC). The persistent inflammatory activity, unpredictable disease course, and cumulative symptom burden of IBD may impact dietary habits.1 Emerging evidence highlights the central role of gut–brain interactions in regulating eating behavior.2 Indeed, patients with IBD frequently perceive diet as a trigger of flares, with 15%-53% attributing relapses to food and up to half modifying their diet accordingly.3,4 While dietary restriction may be adaptive during active disease, many patients persist with restrictive eating behaviors beyond flare periods, which may gradually evolve into eating disorders or avoidant/restrictive food intake disorder (ARFID).
First introduced in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (2013), ARFID is characterized by a persistent failure to meet nutritional or energy requirements, resulting in at least one of the following: significant weight loss, nutritional deficiencies, dependence on enteral feeding or oral supplementation, or marked impairment of psychosocial functioning.5
Unlike anorexia nervosa and bulimia nervosa, which are the most prevalent eating disorders, ARFID is not driven by concerns about body weight or image. Instead, it is typically associated with one or more of the following drivers: avoidance of foods based on sensory properties (“picky eating”); low appetite or limited interest in eating; and fear of adverse consequences such as choking, vomiting, abdominal pain, or bloating.6 Therefore, eating disorders and ARFID should be regarded as distinct diagnostic entities, defined by different diagnostic criteria.
The impact of restrictive eating on nutritional status is of particular concern in IBD, where malnutrition affects 10%-70% of patients globally.7–10 Yet, the consequences of altered eating behaviors extend far beyond nutrition, deeply involving psychological well-being. Restrictive eating can trigger a self-perpetuating cycle in IBD: reduced intake leads to malnutrition, which in turn worsens psychological vulnerability and well-being. In addition, the resulting psychological burden—characterized by body image concerns, social withdrawal, and emotional distress—contributes substantially to disability, limiting independence, social participation, and the ability to work or study.11 Hence, ARFID in IBD should be viewed not only as a nutritional challenge but as a major determinant of long-term disease burden and disability.
Accordingly, in this study, we assessed the risk of altered eating behaviors with a specific focus on ARFID in patients with IBD compared with healthy controls (HC). We further examined if disease onset can be an explanatory of the risk of ARFID and we explored clinical and demographic correlates. Finally, we assessed the impact of ARFID on malnutrition risk and disability.
2. Methods
2.1. Study design and inclusion criteria
This cross-sectional study was conducted at two tertiary IBD centers: the IBD Unit of Federico II University Hospital in Naples and San Raffaele Hospital in Milan, both in Italy.
All consecutive adult patients >18 years old with a confirmed diagnosis of IBD were enrolled between October 2024 and April 2025 during routine outpatient visits and stratified by age at diagnosis. The childhood‐onset cohort included IBD patients diagnosed at ≤18 years of age, and the adulthood‐onset cohort IBD patients diagnosed at >18 years.
Patients were excluded if they had a history of eating disorders or severe unmanaged psychiatric conditions, reported active substance misuse, or were pregnant.
2.2. Data collection
Demographic and clinical data were collected at the time of enrolment, including information on previous treatment exposure, exclusive enteral nutrition (EEN), surgical history, and concomitant extraintestinal manifestations. Disease phenotype was classified according to the Montreal classification.12
Clinical activity was assessed using the partial Mayo score (pMayo) for UC and the Harvey–Bradshaw Index (HBI) for CD. Endoscopic activity was evaluated with the Mayo Endoscopic Subscore (MES) for UC and the Simple Endoscopic Score for Crohn’s Disease (SES-CD) for CD. Clinical activity was defined as pMayo > 2 for UC and HBI > 4 for CD. Endoscopic remission was defined as MES < 1 and SES-CD < 2. Clinical relapse was defined as a composite of change in medical therapy, use of corticosteroids, or IBD-related hospitalization in the previous 12 months. Laboratory findings included C-reactive protein (CRP) (mg/L), fecal calprotectin (µg/g), and hemoglobin (g/dL). Current therapy was also recorded. After receiving standardized training, the investigators (OMN, GC, AMG, FDA, SB) instructed the participants on how to complete the questionnaires, enabling them to do so independently. All responses were collected anonymously and coded to ensure confidentiality. Using the same identification code, a physician independently recorded the disease history score in a separate dataset.
During the same period, HC without a history of gastrointestinal or eating disorders, matched for age, sex, and body mass index (BMI), were recruited among hospital staff, medical students, and visitors or relatives of non-IBD patients.
2.3. Assessment of ARFID risk
ARFID risk was evaluated using the validated Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS-9). The NIAS-9 comprises three domains—Picky eating, Low appetite, and Fear of adverse consequences—each assessed with three questions.13 Each question is rated on a six-point Likert scale (0=“strongly disagree” to 5=“strongly agree”), resulting in a total score between 0 and 45. Specifically, the NIAS-Picky subscale, which assesses sensory-based food avoidance (items 1-3), used a cut-off score of ≥10; the NIAS-Interest subscale, evaluating lack of interest in eating (items 4-6), had a cut-off of ≥9; and the NIAS-Fear subscale, measuring fear of aversive consequences (items 7-9), had a cut-off of ≥10.14 Consistent with prior studies, a cut-off score of ≥24 was used to identify individuals at risk for ARFID.14
2.4. Assessment of eating disorder risk
Eating disorders risk was assessed using the validated Eating Attitudes Test-26 (EAT-26). The EAT-26 consists of 26 items organized into three subscales—dieting, bulimia/food preoccupation, and oral control.15 Each item is rated on a four-point scale ranging from 0 (“never”) to 3 (“always”), yielding a total score between 0 and 78. A cut-off score of ≥20 was used to indicate risk of eating disorders and the need for further clinical evaluation.16
2.5. Assessment of nutritional risk
All patients were assessed for malnutrition with the Patient-Generated Subjective Global Assessment (PG-SGA).17 This tool integrates both patient-reported symptoms (eg, weight changes, dietary intake, nutrition impact symptoms, functional status) and objective clinical findings to provide a more comprehensive evaluation of nutritional status. The overall PG-SGA score ranges from 0 (low risk) to 36 (high malnutrition risk), with a cut-off of ≥6 shown to provide high sensitivity and specificity for identifying IBD patients with malnutrition.18
2.6. Assessment of disability
Disability was assessed using the IBD-Disk, a simplified patient-reported outcome measure derived from the IBD-DI,19 recently validated in Italian.20 The IBD-Disk consists of 10 visual analog scales covering symptoms and functioning domains, displayed graphically as a “disk” to support patient–clinician communication. Disability was considered moderate-to-severe if the IBD-Disk score was ≥40.21 The modified version (mDISK), which includes an additional item assessing dietary habits formulated as: “I have not been able to eat what I want,” rated on a 1–10 scale, was also administered.20 This item was added to specifically capture perceived dietary limitations, which are not explicitly assessed in the original IBD-Disk domains.
2.7. Statistical analysis and ethical considerations
Continuous variables were reported as mean ± standard deviation (SD) or median with interquartile range (IQR), as appropriate, and compared using Student’s t-test or the Mann–Whitney U test. Categorical variables were expressed as frequencies with percentages and compared using the chi-square test or Fisher’s exact test when appropriate. Multivariable analysis was performed using a stepwise binary logistic regression model, including variables that were statistically significant at univariable analysis, as well as variables considered clinically relevant. To avoid multicollinearity, selected variables were excluded from the multivariable model, particularly when composite scores (eg, total IBD-Disk and mDISK) showed strong overlap with related domains. In these cases, only the most informative and non-redundant variables were retained. Subgroup analyses were performed according to age at diagnosis (pediatric-onset vs adult-onset IBD).
All statistical analyses were performed using R software (R Foundation for Statistical Computing, Vienna, Austria), with a two-sided P < .05 considered statistically significant.
The study was conducted in accordance with the “Strengthening the Reporting of Observational Studies in Epidemiology” (STROBE)22 guidelines for cross-sectional studies (Supplementary Table 1). The study was approved by the local Ethics Committee (protocol number: 117.202) and carried out in line with the principles of the Declaration of Helsinki.
3. Results
3.1. Participants’ demographics
Of the 858 individuals invited, 706 completed the questionnaires, yielding a response rate of 82.3%. The study population included 355 patients with IBD (50.3%), nearly half of whom (49.6%) had been diagnosed before age 18, and 351 healthy controls (49.7%). The mean age was 36.3 ± 16.7 years in the IBD group and 35.1 ± 16.2 years in the HC (P = .34). Females accounted for 52.4% of IBD patients and 49.6% of controls (P = .45). Median BMI was slightly lower in the IBD cohort compared with HC (22.4 [IQR 19.6-26.6] vs 22.9 [IQR 20.3-27.1]), although this difference did not reach statistical significance (P = .32). Within the IBD cohort, 176 (49.6%) had CD and 179 (50.4%) UC. At the time of evaluation, 27.6% of IBD patients had clinically active disease, and more than half (55.8%) showed endoscopic activity. The baseline characteristics of the IBD population are summarized in Table 1.
Table 1.
Demographic data of the included population.
| Total population (n = 355) | Pediatric-onset IBD (n = 176) | Adult-onset IBD (n = 179) | P-value | |
|---|---|---|---|---|
| Female sex, n (%) | 186 (52.4) | 93 (52.8) | 93 (52) | .87 |
| Median age, years (IQR) | 30 (22-48) | 22 (22-28) | 47 (34-58) | <.001 |
| Median disease duration, months (IQR) | 108 (48-168) | 120 (72-180) | 96 (48-168) | .01 |
| Previous surgery for IBD, n (%) | 84 (23.7%) | 39 (22.3) | 45 (25.1) | .53 |
| Previous surgery for perianal disease, n (%) | 45 (12.7%) | 28 (15.9) | 17 (9.5) | .07 |
| Previous failure to biological therapy, n (%) | 118 (33.2) | 63 (35.8) | 55 (30.7) | .31 |
| Previous EEN, n (%) | 59 (16.6) | 59 (33.5) | – | <.001 |
| Smoking habits, n (%) | 59 (16.6) | 29 (16.5) | 30 (16.8) | .94 |
| Alcohol consumption, n (%) | 76 (21.4) | 32 (18.2) | 44 (24.6) | .14 |
| Ostomy, n (%) | 10 (2.8) | 3 (1.7) | 7 (3.9) | .20 |
| CD, n (%) | 176 (49.6) | 88 (50.0) | 88 (49.2) | .87 |
| Location | ||||
| Ileal | 54 (30.7) | 15 (17.0) | 39 (44.3) | |
| Colonic | 17 (9.7) | 10 (11.4) | 7 (8.0) | <.001 |
| Ileo-colonic | 105 (59.7) | 63 (71.6) | 42 (47.7) | |
| Behavior | ||||
| Inflammatory | 89 (50.6) | 45 (51.1) | 44 (50.0) | |
| Stricturing | 61 (34.7) | 29 (33.0) | 32 (36.4) | .84 |
| Penetrating | 26 (14.8) | 14 (15.9) | 12 (13.6) | |
| UC, n (%) | 179 (50.4) | 50 (88.0) | 91 (50.8) | .87 |
| Extension | ||||
| Proctitis | 20 (11.2) | 7 (8.0) | 13 (14.3) | |
| Left colitis | 62 (34.6) | 29 (33.0) | 33 (36.3) | .05 |
| Extensive/pancolitis | 97 (54.2) | 52 (59.1) | 45 (49.5) | |
| Perianal disease, n (%) | 48 (27.3) | 33 (44) | 15 (17.0) | |
| Active EIMs, n (%) | ||||
| PA | 18 (5.1) | 16 (9.1) | 2 (1.1) | |
| SpA | 10 (2.8) | 0 | 10 (11.1) | |
| EN | 5 (1.4) | 2 (1.1) | 3 (1.7) | <.001 |
| HS | 5 (1.4) | 4 (2.3) | 1 (0.6) | |
| Psoriasis | 13 (3.7) | 6 (3.4) | 7 (3.9) | |
| CSP | 4 (1.1) | 3 (1.7) | 1 (0.6) | |
| Clinically active disease, n (%) | 98 (27.6) | 53 (30.1) | 45 (25.1) | .30 |
| Endoscopically active disease, n (%) | 198 (55.8) | 88 (50.0) | 110 (61.5) | .03 |
| Median BMI, kg/m2 (IQR) | 22.4 (19.6-26.6) | 21.4 (18.5-24.8) | 23.6 (20.6-28.6) | .22 |
| High malnutrition risk, n (%) | 74 (20.8) | 30 (17.0) | 44 (24.6) | .08 |
| Median Hb, g/dL (IQR) | 13.1 (11.5-14.2) | 13.1 (11.8-14.3) | 12.8 (10.5-13.9) | .11 |
| Median CRP, g/dL | 5.6 (1.7-19.0) | 2.9 (0.5-6.9) | 16.5 (3.5-27.4) | <.001 |
| Median FC, µg/g | 123 (40-370) | 193 (57-560) | 61 (33-200) | <.001 |
| Biological therapy, n (%) | ||||
| IFX | 48 (13.5) | 16 (9.1) | 32 (17.9) | |
| ADA | 85 (23.9) | 47 (26.7) | 38 (21.2) | |
| VDZ | 53 (23.9) | 16 (9.1) | 37 (20.7) | |
| USK | 46 (13.0) | 23 (13.1) | 23 (12.8) | <.001 |
| TOFA | 5 (1.4) | 1 (0.6) | 4 (2.2) | |
| UPA | 13 (3.7) | 4 (2.3) | 9 (5.0) | |
| RISA | 13 (3.7) | 9 (5.1) | 4 (2.2) | |
| Dual therapy | 7 (2.0) | 4 (2.3) | 3 (1.7) |
Abbreviations: IQR, interquartile range; CD, Crohn’s disease; UC, ulcerative colitis; IBD, inflammatory bowel disease; Hb, hemoglobin; CRP, C-reactive protein; FC, fecal calprotectin; BMI, body mass index; EEN, exclusive enteral nutrition; IFX, infliximab; ADA, adalimumab; VDZ, vedolizumab; USK, ustekinumab; TOFA, tofacitinib; UPA, upadacitinib; RISA, risankizumab.
3.2. Risk of eating disorders and ARFID in IBD patients vs HC
The risk of eating disorders (11.3% vs 10.8%) and median EAT-26 scores (6 [IQR 3-12] vs 6 [IQR 2-13]) did not differ between IBD and HC (both P = .91). Instead, the ARFID risk in the IBD population was 13.5% (n = 48), more than double that observed in HC (5.7%, n = 20, P < .001). Importantly, these findings were confirmed also in patients with IBD in clinical remission, who had a significantly increased risk of ARFID compared to HC (10.1% vs 5.7%; P = .03).
Although the median NIAS-9 scores were similar between IBD patients and HC (10 [IQR 4-19] vs 9 [IQR 5-15], P = .17), we observed significant differences within NIAS-9 subdomains: “Fear of adverse consequences” was rated higher in IBD patients compared to HC (3 [IQR 0-7] vs 2 [IQR 0-3], P < .001). Conversely, “Picky eating” scores were lower in IBD patients compared with controls (3 [0-7] vs 4 [2-7], P < .001), as well as “Low appetite” (2 [IQR 0-5] vs 3 [IQR 1-4], P = .03) (Figure 1A).
Figure 1.
NIAS total score and subdomains across clinical subgroups: (A) patients with inflammatory bowel disease (IBD) versus healthy controls (HC); (B) pediatric-onset versus adult-onset IBD; (C) Crohn’s disease (CD) versus ulcerative colitis (UC).
Taken together, these findings indicate that IBD patients are more likely to meet the threshold for ARFID risk, yet the overall risk of general eating disorders does not differ significantly from HC (Figure 2).
Figure 2.
Distribution of avoidant/restrictive food intake disorder (ARFID) domains based on the NIAS-9 among patients with inflammatory bowel disease (IBD) and healthy controls.
When stratifying patients according to age at diagnosis, pediatric-onset and adult-onset IBD showed some distinct baseline characteristics (Table 1). Compared to adult-onset patients, patients with pediatric-onset disease were significantly younger at the time of evaluation (median age 22 [22-28] vs 47 [34-58] years, P < .001), had a longer disease duration (120 [72-180] vs 96 [48-168] months, P = .01), and were less likely to have endoscopic disease activity (61.5% vs 50.0%, P = .03). As expected, history of exclusive enteral nutrition (EEN) was more common in the pediatric-onset group (33.5% vs 0%, P < .001) than in adults, particularly among patients with CD.
The risk of eating disorders did not differ between adults and pediatric-onset patients (10.8% vs 11.7%, P = .46). In terms of ARFID risk, no significant difference was observed between pediatric-onset IBD (21/176, 11.9%) and adult-onset IBD (27/179, 15.1%) (P = .39). The fear of adverse consequences showed no significant difference between the two groups (3 [0-6] vs 3 [0–8], P = .13). A trend toward higher “picky eating” scores was observed in pediatric-onset patients compared with adult-onset, though this was not statistically significant (4 [1-7] vs 3 [0-6], P = .07). Conversely, “low appetite” was more pronounced in adult-onset IBD (3 [0-6] vs 1 [0-4], P = .02) than in pediatric-onset (Figure 1B).
3.3. Factors associated with ARFID risk in IBD patients
At univariable analysis, patients at risk of ARFID were more frequently female (66.7% vs 50.2%, P = .03) and slightly older (median age 32 [24-58] vs 30 [22-47] years, P < .001) compared with those not at risk. Patients with UC showed a trend to an increased ARFID risk, without reaching significant differences with CD (62.5% UC vs 37.5% CD; P = .07). Consistently, no significant differences in NIAS total (8.5 [4.0-16,8] in CD vs 10 [3.25-17.0] in UC; P = .91) or subdomain scores were observed between CD and UC, as shown in Figure 1C.
Patients at risk of ARFID also had a shorter disease duration (102 [39-180] vs 108 [60-168] months, P = .01), fewer previous clinical relapses (0 [0-0.75] vs 0 [0-1], P < .001), and a lower prevalence of alcohol consumption (8.3% vs 23.5%, P = .02). Clinical activity was significantly more common in the ARFID group (45.8% vs 24.8%, P = .002), whereas no differences emerged for endoscopic activity (64.6% vs 54.4%; P = .19) or perianal disease (10.4% vs 14.0%; P = .99). Moreover, ARFID risk was strongly associated with moderate to severe disability (72.9% vs 32.9%, P < .001), dietary impairment on mDISK Q11 (72.9% vs 28.0%, P < .001), malnutrition risk (47.9% vs 16.6%, P < .001), and screening-positive for the risk of other eating disorders (29.2% vs 8.5%, P < .001).
At multivariable logistic regression, independent associated factors of ARFID risk included clinical disease activity (odds ratio [OR] 2.34, 95% CI 1.10-5.01, P = .03), dietary impairment according to mDISK Q11 (OR 4.32, 95% CI 1.92-9.74, P < .001), screen-positive risk for other eating disorders (OR 7.47, 95% CI 2.95-18.90, P < .001), and malnutrition risk (OR 2.31, 95% CI 1.04-5.13, P = .04) (Table 2A, Figure 3A).
Table 2A.
Factors associated with ARFID risk in IBD at univariable and multivariable analysis.
| Overall |
|||||
|---|---|---|---|---|---|
| Univariable analysis |
Multivariable analysis |
||||
| ARFID risk (n = 48) | No ARFID risk (n = 307) | P-value | Odds ratios (95% CI) | P-value | |
| Female sex, n (%) | 32 (66.7) | 154 (50.2) | .03 | 1.35 (0.64-2.85) | .44 |
| Median age, years (IQR) | 32 (24-58) | 30 (22-47) | <.001 | 1.01 (0.99-1.03) | .52 |
| Median BMI, kg/m2 (IQR) | 22.4 (18.9-29.1) | 22.4 (19.7-26.4) | .92 | ||
| CD, n (%) | 18 (37.5) | 158 (51.5) | .07 | ||
| Disease duration, months (IQR) | 102 (39-180) | 108 (60-168) | .01 | 1.00 (0.99-1.01) | .16 |
| Pediatric-onset IBD, n (%) | 21 (43.8) | 155 (50.5) | .39 | ||
| Previous EEN, n (%) | 10 (20.8) | 49 (16.0) | .40 | ||
| Smoking habits, n (%) | 4 (8.3) | 55 (17.9) | .09 | ||
| Alcohol consumption, n (%) | 4 (8.3) | 72 (23.5) | .02 | 0.40 (0.11-1.39) | .15 |
| Previous surgery for IBD, n (%) | 10 (20.8) | 74 (24.1) | .67 | ||
| Perianal disease, n (%) | 5 (10.4) | 43 (14.0) | .99 | ||
| Median n of previous relapses (IQR) | 0 (0-0.75) | 0 (0-1) | <.001 | 0.58 (0.32-1.10) | .07 |
| Clinical activity, n (%) | 22 (45.8) | 76 (24.8) | .002 | 2.34 (1.10-5.01) | .03 |
| Endoscopic activity, n (%) | 31 (64.6) | 167 (54.4) | .19 | ||
| Total IBD-Disk, median (IQR) | 64 (35-78) | 27 (13-46) | .11 | ||
| Total mDISK, median (IQR) | 69 (43-88) | 30 (15-50) | .15 | ||
| mDISK Q 11 | 35 (72.9) | 86 (28.0) | <.001 | 4.32 (1.92-9.74) | <.001 |
| Moderate-to-severe disability, n (%) | 35 (72.9) | 101 (32.9) | <.001 | 1.84 (0.77-4.40) | .17 |
| ED risk, n (%) | 14 (29.2) | 26 (8.5) | <.001 | 7.47 (2.95-18.90) | <.001 |
| Malnutrition risk, n (%) | 23 (47.9) | 51 (16.6) | <.001 | 2.31 (1.04-5.13) | .04 |
Abbreviations: ARFID, avoidant/restrictive food intake disorder; BMI, body mass index; CD, Crohn’s disease; ED, eating disorders; EEN, exclusive enteral nutrition.
Statistically significant P-values are shown in bold.
Figure 3.
Forest plots of multivariable logistic regression analyses identifying factors associated with avoidant/restrictive food intake disorder (ARFID) risk in the overall inflammatory bowel disease (IBD) population (A), pediatric-onset IBD (B), and adult-onset IBD (C).
When stratified according to age of diagnosis, univariable analysis of pediatric-onset patients with ARFID risk showed significantly higher disability scores (IBD-Disk median 68 vs 23, P < .001; mDISK median 74 vs 26, P < .001), greater dietary impairment on mDISK Q11 (71.4% vs 27.1%, P < .001), and a higher prevalence of malnutrition risk (52.4% vs 12.3%, P < .001).
Previous exposure to EEN was comparable between pediatric IBD patients with and without ARFID risk (47.6% vs 31.6%, P = .15). This was also observed in the pediatric-onset CD subgroup, in which EEN use did not differ between patients with and without ARFID risk (20.8% vs 16.0%, P = .40). Consistently, NIAS total and subdomain scores did not differ according to EEN exposure within this subgroup (NIAS total: 9.5 [4-17.3] in EEN exposed vs 7.0 [2-16.3] in EEN non-exposed, P = .28; picky eating: 4 [1-8] vs 3 [0-7], P = .47; appetite: 1 [0-6] vs 1 [0-4], P = .24; fear of aversive consequences: 3 [0-6] vs 2 [0-6.3], P = .34). The multivariable model showed that dietary impairments assessed with mDISK (OR 3.68, 95% CI 1.15-11.73, P = .03) and malnutrition risk (OR 4.39, 95% CI 1.44-13.33, P = .009) were independently associated with ARFID risk (Table 2B, Figure 3B).
Table 2B.
Factors associated with ARFID risk in pediatric-onset IBD patients at univariable and multivariable analysis.
| Pediatric-onset IBD |
|||||
|---|---|---|---|---|---|
| Univariable analysis |
Multivariable analysis |
||||
| ARFID risk (n = 21) | No ARFID risk (n = 155) | P-value | Odds ratios (95% CI) | P-value | |
| Female sex, n (%) | 13 (61.9) | 80 (51.6) | .38 | ||
| Median age, years (IQR) | 23 (21-27) | 22 (20-28) | .94 | ||
| Median BMI, kg/m2 (IQR) | 21.1 (18.6-26.2) | 21.5 (18.4-24.8) | .64 | ||
| CD, n (%) | 9 (42.9) | 79 (60.0) | .49 | ||
| Disease duration, n (%) | 108 (54-168) | 120 (72-180) | .37 | ||
| Previous EEN, n (%) | 10 (47.6) | 49 (31.6) | .15 | ||
| Smoking habits, n (%) | 1 (4.8) | 28 (18.1) | .12 | ||
| Alcohol consumption, n (%) | 2 (9.5) | 30 (19.4) | .27 | ||
| Previous surgery for IBD, n (%) | 4 (19.0) | 35 (22.6) | .79 | ||
| Perianal disease, n (%) | 3 (14.3) | 30 (19.4) | .70 | ||
| Median n of previous relapses (IQR) | 0 (0-1) | 0 (0-2) | .06 | ||
| Clinical activity, n (%) | 9 (42.9) | 44 (28.4) | .18 | ||
| Endoscopic activity, n (%) | 11 (52.4) | 77 (49.7) | .82 | ||
| Total IBD-Disk, median (IQR) | 68 (47-86) | 23 (11-42) | <.001a | ||
| Total mDISK, median (IQR) | 74 (52-95) | 26 (12-46) | <.001a | ||
| mDISK Q 11 | 15 (71.4) | 42 (27.1) | <.001 | 3.68 (1.15-11.73) | .03 |
| Moderate-to-severe disability, n (%) | 17 (81.0) | 47 (30.3) | <.001 | 3.34 (0.88-12.63) | .08 |
| ED risk, n (%) | 4 (19.0) | 15 (9.7) | .19 | ||
| Malnutrition risk, n (%) | 11 (52.4) | 19 (12.3) | <.001 | 4.39 (1.44-13.33) | .009 |
Abbreviations: ARFID, avoidant/restrictive food intake disorder; BMI, body mass index; CD, Crohn’s disease; ED, eating disorders; EEN, exclusive enteral nutrition.
Excluded due to risk of colinearity.
Statistically significant P-values are shown in bold.
In the adult-onset subgroup, ARFID risk was more frequent among women (70.4% vs 48.7%, P = .04) and tended to be associated with older age at evaluation (56 [41–68] vs 46 [33–57] years, P = .06). Patients at risk of ARFID also had a lower prevalence of alcohol consumption (7.4% vs 27.6%, P = .03) and were more likely to present with clinically active disease (48.1% vs 21.1%, P = .003). Those patients presented higher disability scores (IBD-Disk median 46 vs 32, P < .001; mDISK median 51 vs 33, P < .001), greater dietary impairment (74.1% vs 28.9%, P < .001), a higher prevalence of moderate to severe disability (66.7% vs 35.5%, P = .002), and screen-positive risk for other eating disorders (37.0% vs 7.2%, P < .001). Moreover, at multivariable analysis, factors independently associated with ARFID risk in adult-onset IBD included dietary impairment on mDISK Q11 (OR 5.62, 95% CI 1.75-17.99, P = .004) and screen-positive risk for other eating disorders (OR 8.80, 95% CI 2.74-28.30, P < .001) (Table 2C, Figure 3C).
Table 2C.
Factors associated with ARFID risk in adult-onset IBD patients at univariable and multivariable analysis.
| Pediatric-onset IBD |
|||||
|---|---|---|---|---|---|
| Univariable analysis |
Multivariable analysis |
||||
| ARFID risk (n = 27) | No ARFID risk (n = 152) | P-value | Odds ratios (95% CI) | P-value | |
| Female sex, n (%) | 19 (70.4) | 74 (48.7) | .04 | 0.68 (0.24-1.98) | .48 |
| Median age, years (IQR) | 56 (41-68) | 46 (33-57) | .06 | ||
| Median BMI, kg/m2 (IQR) | 24.8 (19.5-29.4) | 23.5 (20.7-28.0) | .61 | ||
| CD, n (%) | 9 (33.3) | 79 (52.0) | .07 | ||
| Disease duration, n (%) | 84 (36-264) | 96 (48-156) | .76 | ||
| Smoking habits, n (%) | 3 (11.1) | 27 (17.8) | .39 | ||
| Alcohol consumption, n (%) | 2 (7.4) | 42 (27.6) | .03 | 0.21 (0.04-1.09) | .06 |
| Previous surgery for IBD, n (%) | 6 (22.2) | 39 (25.7) | .71 | ||
| Perianal disease, n (%) | 2 (7.4) | 13 (8.6) | .65 | ||
| Median n of previous relapses (IQR) | 0 | 0 | .18 | ||
| Clinical activity, n (%) | 13 (48.1) | 32 (21.1) | .003 | 2.50 (0.91-6.82) | .08 |
| Endoscopic activity, n (%) | 20 (74.1) | 90 (59.2) | .14 | ||
| Total IBD-Disk, median (IQR) | 46 (32-74) | 32 (19-49) | <.001a | ||
| Total mDISK, median (IQR) | 51 (39-77) | 33 (20-52) | <.001a | ||
| mDISK Q 11 | 20 (74.1) | 44 (28.9) | <.001 | 5.62 (1.75-17.99) | .004 |
| Moderate-to-severe disability, n (%) | 18 (66.7) | 54 (35.5) | .002 | 1.36 (0.44-4.23) | .60 |
| ED risk, n (%) | 10 (37.0) | 11 (7.2) | <.001 | 8.80 (2.74-28.30) | <.001 |
| Malnutrition risk, n (%) | 12 (44.4) | 32 (21.2) | .009 | 1.22 (0.36-3.47) | .82 |
Abbreviations: ARFID, avoidant restrictive food intake disorder; BMI, body mass index; CD, Crohn’s disease; ED, eating disorders.
Excluded due to risk of colinearity.
Statistically significant P-values are shown in bold.
3.4. Relationship between ARFID risk and malnutrition
Overall, 74 of 355 patients with IBD (20.8%) were at risk of high malnutrition based on PG-SGA ≥ 6. Individuals with ARFID had a significantly higher risk of malnutrition compared with healthy controls (47.9% vs 16.6%, P < .001). At univariable analysis, clinical activity (37.8% vs 24.9%, P = .02), less alcohol consumption (12.2% vs 23.8%, P = .02), ARFID risk (31.0% vs 8.9%; P < .001), and moderate to severe disability (27.4% vs 79.7%, P < .001) were associated with a higher risk of malnutrition. Independent risk factors associated with malnutrition risk were moderate to severe disability (OR = 8.49, CI = 4.4-16.3; P < .001) and ARFID risk (OR = 2.4, CI = 1.2-5.0; P = .014).
3.5. Relationship between ARFID risk and disability
In the IBD cohort, moderate to severe disability was observed in 38.3% of patients, with a median IBD-Disk score of 31 (IQR 16-52). Individuals with ARFID exhibited a significantly greater prevalence of moderate to severe disability compared with HC (72.9% vs 32.9%, P < .001). A higher proportion of moderate to severe disability was found in patients with clinically active disease (38.2% vs 21.0%; P < .001), without alcohol habits (11.0% vs 27.8%; P < .001), at risk of malnutrition (43.4% vs 6.8%; P < .001), and ARFID (25.7% vs 5.9%; P < .001). Moreover, patients with moderate to severe disability had a shorter disease duration (84 [48-153] vs 120 [60-192]; P = .007). At multivariable analysis, alcohol consumption was inversely associated with moderate to severe disability (OR 0.41, 95% CI 0.21-0.82, P = .012). In contrast, clinical disease activity significantly increased the odds of moderate to severe disability (OR 1.90, 95% CI 1.09-3.20, P = .02). Malnutrition risk emerged as the strongest factor, with more than a 10-fold higher risk (OR 8.60, 95% CI 4.44-16.64, P < .001). Similarly, ARFID risk increased the odds of moderate to severe disability (OR = 3.20, 95% CI 1.49-6.88; P = .003). Finally, longer disease duration was slightly but significantly protective (OR 0.997, 95% CI 0.994-0.999, P = .013).
4. Discussion
Patients with IBD often modify their diet in an attempt to control symptoms. However, these adaptive disorders can extend beyond limits and evolve into altered eating behaviors. Self-restrictive diets and ARFID, especially if followed without professional guidance, can have negative effects on both nutritional and psychological aspects of health. To our knowledge, this is the first study to assess the prevalence of eating disorders and ARFID risks among individuals with pediatric- and adult-onset IBD compared to the general population. Our a priori hypotheses were that: (1) ARFID symptoms are common among individuals with IBD; (2) age at IBD onset is a factor associated with eating disorders and ARFID risk; (3) altered eating disorders and restrictive behaviors negatively affect nutritional status; and (4) individuals with IBD who meet screening criteria for ARFID experience greater disability than those without ARFID.
Distinguishing adaptive food avoidance in IBD from pathological restriction remains complex. However, structured screening plays a key role in guiding healthcare providers toward focused clinical questioning, supporting early identification of at-risk individuals, and identifying those who may benefit from targeted nutritional or psychological assessment.
In the present study, the overall risk of eating disorders, as measured by the EAT-26, was similar between patients and controls. Conversely, the risk of ARFID was significantly higher in IBD (13.5% vs. 5.6%) than in HC, with no significant difference between CD and UC. Notably, NIAS-9 subdomains revealed that fear of adverse consequences from eating was more pronounced in IBD patients compared with controls. Taken together, this suggests that eating behaviors in IBD are mainly driven by ARFID-related mechanisms—fear of symptom exacerbation, appetite suppression related to disease activity, and food avoidance—rather than by classical psychiatric eating disorders such as anorexia nervosa or bulimia, which are more strongly rooted in psychological and sociocultural factors.
Notably, we also investigated the impact of age at disease onset on the risk of eating disorders and ARFID, as onset during vulnerable developmental stages may affect psychological and psychosocial dimensions. Indeed, a recent retrospective observational study showed that children and young adults with incident IBD have a higher incidence (adjusted hazard ratio of 1.85) of new mental health conditions compared with controls.23 However, in our analysis the prevalence of ARFID and eating disorders risk was comparable between pediatric and adult onset, suggesting that adaptive responses to chronic disease at pediatric age may converge over time, leading to a uniform pattern of eating behaviors across age groups. We did not observe a significant association between EEN exposure and ARFID risk in our pediatric cohort, although prior evidence suggests that exclusion diets in CD may reinforce food-related fears.24 Given that this analysis was exploratory, not specifically designed to assess this relationship, and potentially underpowered, this finding should be interpreted with caution, and a potential association cannot be ruled out.
Furthermore, we observed a higher risk of ARFID symptoms in females with IBD compared to HC (66.7% vs 50.2%). This is consistent with the broader epidemiological trend of eating disorders, which have higher prevalence among females.25 However, in our cohort, female sex alone was not associated with ARFID, suggesting that disease-related factors such as activity of disease may play a more decisive role in the development of restrictive eating behaviors. Accordingly, our data confirmed a strong association between active disease and the risk of ARFID. Patients with clinically active IBD were more likely to screen positive for ARFID, and those at risk of ARFID had higher rates of malnutrition. This is consistent with previous studies showing that ARFID behaviors are particularly common in patients with active gastrointestinal symptoms, often as part of self-directed, unsupervised dietary strategies.14,26 However, even in IBD patients in clinical remission, ARFID risk is not neglectable compared to HC, suggesting that altered change in eating behaviors may persist beyond the resolution of intestinal inflammation. This has been confirmed in a recent study27 showing that the rate of ARFID based on DSM-5 criteria among patients with inactive disease was 16.3%, which was not statistically different from the rate observed in those with active disease (23.4%).
As expected, altered eating behaviors impact on nutritional status: nearly half of IBD patients at risk of ARFID in our cohort were also at risk of malnutrition. Indeed, malnutrition and weight loss are part of formal DSM-5 ARFID diagnostic criteria. In the present study, however, these parameters were treated as correlates of ARFID risk rather than diagnostic determinants. Given the established link between malnutrition and adverse IBD outcomes, including its recognition as one of the 10 key quality outcome indicators for IBD, screening for dietary habits should be part of the routine clinical assessment.28,29
Restrictive eating behaviors may also adversely influence activity limitations and participation restrictions, thereby contributing to disability. Our study revealed a strong association between disability and ARFID, confirming functional consequences beyond nutritional status alone. However, the current IBD-Disk does not cover diet habits. As such, we have recently developed a modified version of the IBD-Disk20 that incorporates a domain on diet-related impairment. Notably, in our analysis, dietary impairment assessed by the mDISK independently explained ARFID risk, suggesting that it could be easily integrated into routine clinical practice as a practical early screening tool and help identify patients requiring psychological support.
The strengths of our study include the large sample size and the use of validated screening questionnaires. Unlike most studies limited to IBD cohorts, our study included an age- and sex-matched control group from the general population that improves the generalizability of our results. Importantly, eating disorders and ARFID are distinct diagnostic entities, and prevalence estimates derived from one instrument do not necessarily reflect the other, as they capture different dimensions of eating behaviors. To account for this, we employed distinct screening questionnaires. The NIAS identifies characteristic patterns of restrictive eating but does not establish a diagnosis or differentiate between behaviors driven by organic disease and behaviors consistent with ARFID. Yet it has shown high internal consistency, acceptable test–retest reliability, and good convergent and discriminant validity in ARFID risk detection, supporting its use as a screening rather than diagnostic tool.13,30 The EAT-26 has been widely used in both pediatric and adult populations, including individuals with IBD, and is recognized as a reliable screening tool for eating disorders.31 Although these instruments target different constructs, individuals may screen positive for ARFID risk and other eating-disorder risk, indicating overlapping screening profiles rather than coexisting diagnoses.
We acknowledge that in the context of an underlying medical condition, altered eating behaviors may be shaped or worsened by the diagnosis itself and can represent physical symptoms or adaptive responses rather than independent pathological patterns. However, an underlying medical diagnosis should not deter clinicians from investigating ARFID risk. Importantly, emerging evidence across digestive diseases demonstrates that ARFID and ARFID-like traits occur in patients with diverse organic conditions, including eosinophilic esophagitis, celiac disease, and motility disorders.14,32,33 These data reinforce that organic disease does not preclude the presence of ARFID risk and should not lead to diagnostic overshadowing. Instead it may heighten the need for a comprehensive evaluation that integrates behavioral, nutritional, and psychological dimensions.
Our study also provides novel insights by comparing pediatric- and adult-onset IBD. Notably, age at disease onset did not increase the risk of eating disorders and ARFID. This implies that these risks in IBD are not only attributable to developmental or psychosocial vulnerabilities in younger patients but primarily influenced by disease-related factors, including activity, nutritional status, and disability. Nevertheless, even among patients in clinical remission, the risk of ARFID remained higher than in HC (10.1% vs. 5.7%). This supports the interpretation that ARFID behaviors cannot be fully explained by IBD disease activity. Consequently, systematic screening for ARFID should be implemented in all individuals irrespective of disease activity or age at onset.
The study has some limitations. First, the referral center setting may reduce generalizability, as patients in tertiary care often present with more complex disease and may not reflect the broader IBD population. Second, data on diet, dietitian supervision, or nutritional follow-up were not systematically collected in this study and we are therefore unable to examine differences between patients who were or were not followed by dietitians. Dietitian involvement may influence restrictive behaviors and ARFID risk, and distinguishing between self-initiated dietary changes and supervised dietary recommendations would provide valuable context. However, nutritional care pathways vary widely across centers; access to dietitians, referral criteria, and the intensity of follow-up differ substantially between individuals. As a result, it was difficult to capture reliable and comparable data across the study population. We collected data only on EEN, as it is the most common and validated nutritional therapy particularly in pediatric CD.
Furthermore, the lack of data on socioeconomic status and educational background is relevant, since these factors can shape dietary habits, health literacy, and access to support services, thereby influencing eating behaviors. Future multicenter studies including more diverse populations and incorporating multiple psychosocial variables such as depression and anxiety disorders are warranted to validate and extend these findings.
Conclusions
The risk of ARFID, particularly food-related fear, is nearly twice as high in IBD compared with HC and is closely associated with disease activity and malnutrition. Yet, even in remission, ARFID risk remained elevated compared to HC. The impact of ARFID risk goes beyond the malnutrition, extending to potential harm to emotional well-being, mental health, and long-term disability. Early identification together with patient education on the consequences of maladaptive eating patterns is crucial. Simple tools such as the IBD-Disk may facilitate routine screening and guide timely referral to specialized care. Integrating such instruments into daily practice may reduce underdiagnosis and potentially improve long-term patient outcomes.
Supplementary Material
Contributor Information
Olga Maria Nardone, Gastroenterology, Department of Public Health, University of Naples Federico II, Naples, Italy.
Ferdinando D’Amico, Gastroenterology and Endoscopy, IRCCS San Raffaele Hospital and Vita Salute San Raffaele University, Milan, Italy.
Giulio Calabrese, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Tommaso Lorenzo Parigi, Gastroenterology and Endoscopy, IRCCS San Raffaele Hospital and Vita Salute San Raffaele University, Milan, Italy; Department of Pathology, Case Western Reserve University, Cleveland, OH, United States.
Alfredo Marco Gargiulo, School of Medicine, University of Naples Federico II, Naples, Italy.
Sarah Bencardino, Gastroenterology and Endoscopy, IRCCS San Raffaele Hospital and Vita Salute San Raffaele University, Milan, Italy.
Flavia Palumbo, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Martina Petolicchio, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Alessia Dalila Guarino, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Antonio Rispo, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Anna Testa, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Silvio Danese, Gastroenterology and Endoscopy, IRCCS San Raffaele Hospital and Vita Salute San Raffaele University, Milan, Italy.
Fabiana Castiglione, Gastroenterology, Department of Clinical Medicine and Surgery, University of Naples Federico II, Naples, Italy.
Author contributions
O.M.N.: Conceptualization, Methodology, Writing—Original Draft, Writing—Review & Editing. F.D.A.: Data Curation, Writing—Original Draft, Writing—Review & Editing. G.C.: Conceptualization, Methodology, Formal analysis, Writing—Original Draft, Writing—Review & Editing. T.L.P.: Writing—Original Draft, Writing—Review & Editing. A.M.G.: Data Curation, Writing—Review & Editing. S.B.: Data Curation, Writing—Review & Editing. F.P.: Data Curation, Writing—Review & Editing. M.P.: Data Curation, Writing—Review & Editing. A.D.G.: Writing—Review & Editing. A.R.: Writing—Review & Editing. A.T.: Writing—Review & Editing. S.D.: Supervision, Validation. F.C.: Supervision, Validation.
Supplementary material
Supplementary material is available at ECCO-JCC online.
Funding
None.
Conflicts of interest
O.M.N. reports lecture fees from Ferring, AbbVie, Janssen, Pfizer, Eli Lilly, Alfa Sigma, Noòs, Recordati; Advisory Board from Nestle, Janssen, Eli Lilly. F.D.A. has served as a speaker for AbbVie, AnaptysBio, Ferring, Fresenius Kabi, Sandoz, Janssen, Galapagos, Giuliani, Lilly, Takeda, Tillotts, and Omega Pharma; he has also served as an advisory board member for Ferring, Fresenius Kabi, Galapagos, AbbVie, Janssen, Lilly, Takeda, and Nestlè. T.L.P. has received speaker fees from Abbvie, Takeda, Tillots, Fresenius Kabi, Ferring, and Janssen; Travel grants from Abbvie, Alfasigma, Janssen, CADI group, Pfizer, Takeda. A.R. has served as advisory board for Abbvie, MSD, Takeda, Janssen, Pfizer. A.T. has served as advisory board for Abbvie, Takeda, Janssen. S.D. has served as a speaker, consultant and advisory board member for Schering-Plow, Abbott (AbbVie) Laboratories, Merck, UCB Pharma, Ferring, Cellerix, Millenium Takeda, Nycomed, Pharmacosmos, Actelion, Alfa Wasserman, Genentech, Grunenthal, Pfizer, AstraZeneca, Novo Nordisk, Cosmo Pharmaceuticals, Vifor and Johnson and Johnson. F.C. has served as advisory board for Abbvie, MSD, Takeda, Janssen, Pfizer. All other authors have no disclosures to declare.
Data Availability
Data are available upon reasonable request to the corresponding author.
References
- 1. Limketkai BN, Hamideh M, Shah R, Sauk JS, Jaffe N. Dietary patterns and their association with symptoms activity in inflammatory bowel diseases. Inflamm Bowel Dis. 2022;28:1627-1636. [DOI] [PubMed] [Google Scholar]
- 2. Sirufo MM, Magnanimi LM, Ginaldi L, De Martinis M. Anorexia nervosa and autoimmune comorbidities: a bidirectional route? CNS Neurosci Ther. 2022;28:1921-1929. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Verdina M, Seibold F, Grandmaison G, et al. Survey of dietary beliefs and habits of inflammatory bowel disease patients. Clin Nutr ESPEN. 2023;57:624-629. [DOI] [PubMed] [Google Scholar]
- 4. Crooks B, McLaughlin J, Limdi J. Dietary beliefs and recommendations in inflammatory bowel disease: a national survey of healthcare professionals in the UK. Frontline Gastroenterol. 2022;13:25-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. American Psychiatric Association; 2013. [cited 2025 Dec 12]; Available from: https://psychiatryonline.org/doi/book/10.1176/appi.books.9780890425596
- 6. Hog L, Dinkler L. Recent insights into the epidemiology of avoidant/restrictive food intake disorder (ARFID). Curr Opin Psychiatry. 2025;38:402-409. https://journals.lww.com/10.1097/YCO.0000000000001041 [DOI] [PubMed] [Google Scholar]
- 7. Nguyen GC, Munsell M, Harris ML. Nationwide prevalence and prognostic significance of clinically diagnosable protein-calorie malnutrition in hospitalized inflammatory bowel disease patients. Inflamm Bowel Dis. 2008;14:1105-1111. [DOI] [PubMed] [Google Scholar]
- 8. Casanova MJ, Chaparro M, Molina B, et al. Prevalence of malnutrition and nutritional characteristics of patients with inflammatory bowel disease. J Crohns Colitis. 2017;11:1430-1439. [DOI] [PubMed] [Google Scholar]
- 9. Filippi J, Al-Jaouni R, Wiroth JB, Hébuterne X, Schneider SM. Nutritional deficiencies in patients with Crohnʼs disease in remission. Inflamm Bowel Dis. 2006;12:185-191. [DOI] [PubMed] [Google Scholar]
- 10. Vadan R, Gheorghe LS, Constantinescu A, Gheorghe C. The prevalence of malnutrition and the evolution of nutritional status in patients with moderate to severe forms of Crohn’s disease treated with infliximab. Clin Nutr. 2011;30:86-91. [DOI] [PubMed] [Google Scholar]
- 11. Nardone OM, Calabrese G, La Mantia A, Caso R, Testa A, Castiglione F. Insights into disability and psycho-social care of patients with inflammatory bowel disease. Front Med (Lausanne). 2024;11:1416054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Maaser C, Sturm A, Vavricka SR, et al. ; European Crohn’s and Colitis Organisation [ECCO] and the European Society of Gastrointestinal and Abdominal Radiology [ESGAR]. ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 1: initial diagnosis, monitoring of known IBD, detection of complications. J Crohns Colitis. 2019;13:144-164. [DOI] [PubMed] [Google Scholar]
- 13. Zickgraf HF, Ellis JM. Initial validation of the Nine Item Avoidant/Restrictive Food Intake disorder screen (NIAS): a measure of three restrictive eating patterns. Appetite. 2018;123:32-42. [DOI] [PubMed] [Google Scholar]
- 14. Yelencich E, Truong E, Widaman AM, et al. Avoidant restrictive food intake disorder prevalent among patients with inflammatory bowel disease. Clin Gastroenterol Hepatol. 2022;20:1282-1289.e1. [DOI] [PubMed] [Google Scholar]
- 15. Garner DM, Olmsted MP, Bohr Y, Garfinkel PE. The Eating Attitudes Test: psychometric features and clinical correlates. Psychol Med. 1982;12:871-878. [DOI] [PubMed] [Google Scholar]
- 16. Stein D, Spivak-Lavi Z, Tzischinsky O, Peleg O, Dikstein H, Latzer Y. Differences in the factor structure of the Eating Attitudes Test-26 in female adolescent patients with eating disorders before and after treatment. J Eat Disord. 2025;13:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Jager-Wittenaar H, Ottery FD. Assessing nutritional status in cancer: role of the Patient-Generated Subjective Global Assessment. Curr Opin Clin Nutr Metabolic Care. 2017;20:322-329. [DOI] [PubMed] [Google Scholar]
- 18. Gabrielson DK, Scaffidi D, Leung E, et al. Use of an Abridged Scored Patient-Generated Subjective Global Assessment (abPG-SGA) as a nutritional screening tool for cancer patients in an outpatient setting. Nutr Cancer. 2013;65:234-239. [DOI] [PubMed] [Google Scholar]
- 19. Ghosh S, Louis E, Beaugerie L, et al. Development of the IBD Disk: a visual self-administered tool for assessing disability in inflammatory bowel diseases. Inflamm Bowel Dis. 2017;23:333-340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Nardone OM, Bruzzese D, Allocca M, et al. ; IBD-Disk Italian Study Group. Italian validation of the IBD-disk tool for the assessment of disability in inflammatory bowel diseases: A cross-sectional multicenter study. Dig Liver Dis. 2025;57:753-761. [DOI] [PubMed] [Google Scholar]
- 21. Tadbiri S, Nachury M, Bouhnik Y, et al. ; GETAID-IBD-disk study group. The IBD-disk is a reliable tool to assess the daily-life burden of patients with inflammatory bowel disease. J Crohns Colitis. 2021;15:766-773. [DOI] [PubMed] [Google Scholar]
- 22. Elm EV, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335:806-808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Cooney R, Tang D, Barrett K, Russell RK. Children and young adults with inflammatory bowel disease have an increased incidence and risk of developing mental health conditions: a UK population-based cohort study. Inflamm Bowel Dis. 2024;30:1264-1273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Sigall Boneh R, Westoby C, Oseran I, et al. The Crohn’s disease exclusion diet: a comprehensive review of evidence, implementation strategies, practical guidance, and future directions. Inflamm Bowel Dis. 2024;30:1888-1902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Rangel Paniz G, Lebow J, Sim L, Lacy BE, Farraye FA, Werlang ME. Eating disorders: diagnosis and management considerations for the IBD practice. Inflamm Bowel Dis. 2022;28:936-946. [DOI] [PubMed] [Google Scholar]
- 26. Marsh A, Kinneally J, Robertson T, Lord A, Young A, Radford –Smith G. Food avoidance in outpatients with inflammatory bowel disease – who, what and why. Clin Nutr ESPEN. 2019;31:10-16. [DOI] [PubMed] [Google Scholar]
- 27. Grossberg LB, Mishra K, Rabinowitz LG, et al. A Multicenter study to assess avoidant/restrictive food intake disorder in patients with inflammatory bowel disease. Inflamm Bowel Dis. 2025;31:2381-2389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Melmed GY, Siegel CA. Quality improvement in inflammatory bowel disease. Gastroenterol Hepatol (N Y). 2013;9:286-292. [PMC free article] [PubMed] [Google Scholar]
- 29. Fiorino G, Caprioli FA, Onali S, et al. Adaptation of the European Crohn’s Colitis Organisation quality of care standards to Italy: The Italian Group for the study of inflammatory bowel disease consensus. Dig Liver Dis. 2025;57:1135-1140. [DOI] [PubMed] [Google Scholar]
- 30. Jager-Wittenaar H, Ottery FD. Assessing nutritional status in cancer: role of the Patient-Generated Subjective Global Assessment. Curr Opin Clin Nutr Metabolic Care. 2017;20:322-329. [DOI] [PubMed] [Google Scholar]
- 31. Vickers M, Whitworth J, Alvarez LM, Bowden M. Disordered eating behaviors in pediatric patients with inflammatory bowel disease in remission or mild‐moderate disease activity. Nutr Clin Pract. 2024;39:881-887. [DOI] [PubMed] [Google Scholar]
- 32. Fink M, Simons M, Tomasino K, Pandit A, Taft T. When is patient behavior indicative of Avoidant Restrictive Food Intake Disorder (ARFID) Vs reasonable response to digestive disease? Clin Gastroenterol Hepatol. 2022;20:1241-1250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Flack R, Brownlow G, Burton-Murray H, Palsson O, Aziz I. The prevalence and burden of avoidant/restrictive food intake disorder symptoms in adults with disorders of gut-brain interaction: a population-based study. Gastroenterology. 2026;170:365-374. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request to the corresponding author.