ObjectiveCognitive‐behavioral therapy for avoidant/restrictive food intake disorder (ARFID; CBT‐AR) shows promise in improving clinical outcomes in children/adolescents and adults. We aimed to identify predictors of outcomes in CBT‐AR. We hypothesized that younger age, non‐underweight status, and presence of the fear of aversive consequences profile of ARFID would predict greater likelihood of remission post–treatment, and that presence of the lack of interest in eating/food and sensory sensitivity profiles would predict greater likelihood of persistence post‐treatment. We included sex as an exploratory predictor.MethodIndividuals (N = 94, ages 10–55 years) were offered 20–30 outpatient sessions of CBT‐AR. We collected clinical and demographic data at pre‐treatment, and remission status at post‐treatment.ResultsConsistent with our hypothesis, presence (versus absence) of the fear profile predicted an almost three‐fold increased likelihood of remission. Presence of the sensory profile, lack of interest profile, age, weight status, and sex were not predictors of ARFID outcome.DiscussionThe fear of aversive consequences profile of ARFID may be more amenable to treatment with CBT‐AR. This is the first study to identify predictors of treatment outcome following CBT‐AR. Randomized controlled trials are needed to confirm these findings and examine moderators.