ABSTRACTAnxiety sensitivity (AS), or the tendency to appraise physical symptoms as intolerable or dangerous, may maintain the cycle between co-occurring anxiety and substance use disorders. This study examined preliminary efficacy of a brief intervention targeting AS for individuals with heterogeneous substance use disorders. Forty-one patients with high AS entering an addictions day program were randomized to treatment as usual (TAU) or to TAU plus a nine-hour AS-focused intervention that consisted of interoceptive exposures, psychoeducation about the cycle of problematic substance use and anxiety, and a single session of cognitive challenging (e.g., reviewing common cognitive distortions and decatastrophizing anxiety symptoms). Mixed-effects intent-to-treat models suggested that participants in the AS condition showed greater decreases in AS at post-treatment, but this effect was lost at follow-up three months later. Intervention conditions did not differ in change in percent days abstinent or self-reported anxiety, with both conditions showing significant improvement at post-treatment. Results suggest that the nine-hour AS-focused intervention led to a short-term benefit over TAU alone, but this benefit was not sustained at three months’ follow-up. Future AS interventions may need to target specific subconstructs of AS for selected populations, or target emotional distress tolerance more broadly.