To determine if maintenance CBT produces a more sustained improvement among patients with panic disorder (PD) who respond to an initial course of cognitive behavior therapy (CBT) alone. For those who do not respond sufficiently to CBT alone, to determine if the addition of pharmacotherapy is acceptable and if this improves response among those inadequate responders to CBT alone.

This multicenter study builds upon the findings of a prior study comparing imipramine, placebo, CBT, and their combination in the treatment of PD patients with no more than mild agoraphobia. That study found response rates were as high with CBT or imipramine alone as with their combination. Given the added cost of combined treatments, it therefore seems reasonable to begin with monotherapy. Further, following general principles of medical practice, it would be reasonable to initiate treatment with the less invasive cognitive behavioral intervention. It is then important to learn what should be done following initial treatment.

All patients initially receive CBT alone. Patients are then randomized into 1 of 2 post-acute studies, depending on response status. Responders are randomized to a maintenance study comparing no maintenance with 9 months of continued CBT. Nonresponders are randomized to a study comparing paroxetine with continued CBT. The following outcomes will be examined: the necessity of maintenance therapy in maintaining response; the ability of adjunct pharmacotherapy to improve the response of patients who did not respond to CBT alone; possible predictors of response and relapse; and possible mediators of response.