Effectiveness of Virtually Delivered Cognitive-Behavioral Therapy for Avoidant/Restrictive Food Intake Disorder

Cognitive-behavioral therapy for ARFID (CBT-AR) has demonstrated preliminary efficacy in open trials, but no studies have examined its effectiveness in real-world settings. We evaluated the effectiveness of virtually delivered CBT-AR for adults with ARFID and examined whether baseline symptom severity within each ARFID profile predicted corresponding symptom improvement.

N = 549 adults with ARFID received CBT-AR through Equip Health. We collected weight and Pica, ARFID, and Rumination Disorder Interview-ARFID Questionnaire data at intake and throughout treatment. Ordinary least squares regressions examined changes in weight and ARFID symptoms, and whether baseline severity within each profile predicted improvement in that domain, controlling for severity in other profiles and demographic/clinical covariates.

Patients requiring weight restoration (n = 243) gained an average of 0.34 lb (0.15 kg) per week (p < 0.001). In the full sample, patients demonstrated significant reductions in sensory sensitivity (b = -0.05, 95% CI [-0.07, -0.04]), fear of aversive consequences (b = -0.04, 95% CI [-0.05, -0.02]), and lack of interest in food/eating (b = -0.04, 95% CI [-0.05, -0.03]; all ps < 0.001). Greater baseline severity within each profile uniquely predicted greater improvement in the corresponding profile (all bs = -0.03, all ps < 0.001).

CBT-AR delivered virtually under naturalistic conditions is effective for adults with ARFID, producing significant improvements in weight and ARFID symptoms. Though randomized controlled trials are needed to establish definitive efficacy, findings provide preliminary support for dissemination beyond specialized research settings and demonstrate the clinical utility of profile-based treatment planning.