PREVALENCE AND CHARACTERISTICS OF AVOIDANT/RESTRICTIVE FOOD INTAKE DISORDER IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE: A MULTISITE STUDY

Background: Patients with inflammatory bowel disease (IBD) often modify their diet in attempt to manage their disease and symptoms. In some cases, dietary changes involve excessive restriction, which may place patients at risk for adverse consequences. Avoidant Restrictive Food Intake Disorder (ARFID) is a non-body image based eating disorder that involves restrictive eating and has been suggested to disproportionately affect patients with IBD. However, true rates of ARFID are not yet known, as no study to date has applied ARFID diagnostic criteria. The aims of the current study were (1) to identify the prevalence of IBD patients screening positive for ARFID and (2) explore associations between psychosocial and disease-related factors with ARFID symptoms.
Methods: Adult patients with confirmed IBD were invited to complete electronic questionnaires at two academic medical centers. The ARFID subscale of the Pica, ARFID, and rumination disorder questionnaire (PARDI-AR-Q) was used to categorize patients as screening positive for ARFID. The nine-item ARFID Screen (NIAS) was used to measure ARFID symptom severity. Additional survey items included demographic and disease information, the Visceral Sensitivity Index (VSI), Patient Health Questionnaire-4 (PHQ-4) for anxiety and depression, and the NIH-PROMIS Global Health Questionnaire for mental and physical health. The Harvey Bradshaw Index (HBI) and Patient-Reported Outcome (PRO-2) were used to characterize disease activity in Crohn’s disease (CD) and ulcerative colitis (UC), respectively. Descriptive statistics and Pearson’s correlations were used to analyze sample data.
Results: A total of 258 patients (55.8% female, mean age 47.6 (17)) responded to the survey, of which 136 (53%) had CD and 122 (47%) had UC. 40% of the sample had active disease. ARFID scores are presented in table 1. 17% of the sample screened positive for ARFID with significantly more patients with CD screening positive compared to UC (23.4% vs. 11.1% respectively, p = .012). Patients with active disease were more likely to screen positive for ARFID compared to inactive disease (25.3% vs. 12.4% respectively, p = .011). Higher ARFID scores based on NIAS were associated with higher scores on PHQ-4 anxiety (r = .13, p < .05), VSI (r = .47, p < .001), and PROMIS mental and physical health (r = .30, p < .001 and r = .30, p < .001 respectively).
Conclusions: 17% of tertiary care IBD patients in this multisite study screened positive for ARFID based on a validated screening questionnaire that corresponds with DSM-5 criteria. Disease type and severity placed patients at higher risk for ARFID. Greater ARFID symptomatology was associated with more GI-specific anxiety and quality of life impairment. These findings highlight the need for adequate screening, education, and treatment to prevent and manage ARFID in patients with IBD.