A brief group-based cognitive behavioral therapy (CBT), with running as an interoceptive exposure (IE) component, was effective in reducing anxiety sensitivity (AS) levels in undergraduate women (Watt, Stewart, Lefaivre, & Uman, 2006). This study investigated whether the CBT/IE intervention would result in decreases in AS and emotional distress that would be maintained over 14 weeks. Female undergraduates, high (n = 81) or low (n = 73) in AS, were randomized to 3-day CBT plus forty-two 10-min running IE trials (n = 83) or 3-day health education control (HEC) with interactive discussions and problem solving on exercise, nutrition, and sleep (n = 71). The CBT/IE intervention led to decreases in AS, depression, and stress symptoms for high AS participants, which were maintained at 14 weeks. Unexpectedly, HEC participants experienced similar and lasting decreases in AS, depression, and anxiety symptoms. Furthermore, there were no post-intervention differences between CBT/IE and HEC participants in any of the outcomes. Low AS participants experienced few sustained changes. Clinical implications and the possible role of aerobic exercise in explaining outcomes of both interventions are discussed.