Abstract

The relative effectiveness of the available treatments for panic disorder may best be understood in the context of the longitudinal course of the disorder. This study examines a number of clinically relevant issues, including long-term outcome after acute treatment, the proportion of patients remaining on single-agent treatment or requiring multiple medications or nonpharmacologic interventions over time, evidence for dose escalation during maintenance high-potency benzodiazepine therapy, and predictors of acute and long-term response to treatment. Fifty-nine panic disorder patients originally randomized to treatment in a controlled trial comparing alprazolam, clonazepam, and placebo were reevaluated in a follow-up study. At a mean follow-up of 1.5 years, 78% of patients remained on medication and the mean dosage of alprazolam and clonazepam did not increase. Our data suggest that most patients maintain benefit with long-term pharmacotherapy but that residual symptomatology may require more intensive or additional treatment strategies. Response at the endpoint of the acute trial was significantly associated with pretrial baseline Clinical Global Impression Scale score and the presence of dysthymia. Poor outcome at follow-up was associated with total duration of the disorder, agoraphobic subtype, and the presence of comorbid social phobia. We underscore the potential importance of comorbid affective and anxiety disorders as well as phobic patterns in determining long-term response to treatment.