Patients with panic disorder are a highly symptomatic, help-seeking group who tend toward recurrent exacerbations of symptoms (Pollack and Otto 1997; Pollack and Marzol, 2000; Faravelli et al., 1995). It is therefore important to gauge not only the effectiveness of treatments over the short term but to ascertain their effectiveness over longer follow-up intervals. Useful data with regard to long-term outcome, however, has been limited thus far in the literature. In a review of follow-up studies to date, Milrod et al. (1996) found that most did not monitor concurrent nonstudy treatments (e.g., untracked medication use in CBT studies or ongoing psychotherapies in medication studies) either during study treatment or during follow-up intervals. The authors concluded that there was limited evidence that patients responding to short-term treatments maintained their gains if they did not receive further treatment. Bakker et al. (1998) conducted a meta-analysis of studies that had information on long-term follow-up of treatments of panic disorder. Their analysis indicated that treatments of panic disorder show effectiveness acutely and at follow-up, finding an advantage to antidepressants plus in vivo exposure over cognitive behavioral treatments. Nevertheless, the authors note that an important limitation of their findings was "the studies presented surprisingly little information on the type of additional treatment received, frequency of visits to therapists, and kind and amount of medication during the follow-up period" (p. 417).
Barlow et al.'s (2000) study makes a strong case for the importance of monitoring both acute and longer-term response to treatments in determining the treatment efficacy for panic disorder. Patients in this study received either imipramine, up to 300 mg/d, CBT, CBT plus imipramine, CBT plus placebo, or placebo for a 3-month period. Patients who responded to treatment were seen monthly in a maintenance schedule that provided treatment similar to what they received in the acute phase for an additional 6 months. They were then followed for 6 months after all treatment was discontinued.
After the acute phase, while all active treatments were superior to placebo, CBT plus imipramine showed limited superiority over CBT alone, but not CBT plus placebo. However, CBT plus imipramine was superior to both CBT alone and CBT plus placebo after the maintenance phase. In responders to the treatment, the level of response to imipramine was found to be superior to that of CBT after the acute phase, and this trend continued after the maintenance phase. However, in the follow-up period, responders to imipramine, with or without CBT, had much higher rates of relapse than those who received CBT without medication. At follow-up, CBT alone and CBT plus placebo were superior to placebo, but imipramine and imipramine plus CBT were not. The study suggests that while medication may be a more effective initial intervention, CBT may have a more durable effectiveness, and medication may lead to a greater potential for relapse at drug discontinuation.
More recent studies are being designed more carefully to assess the nature of specific interventions the patients may receive in the follow-up period (Milrod et al., 2001; de Beurs et al., 1999; Mavissakalian and Perel, 1999). De Beurs et al. (1999), for example, in a naturalistic follow-up at 2 years of panic patients treated in four treatment conditions, more carefully assessed and reported interim treatment. Patients had received either fluvoxamine combined with exposure, placebo plus exposure, psychological panic management plus exposure, or exposure alone. Fluvoxamine plus exposure was found to be superior after acute treatment, but there were no significant differences between treatments at 2-year follow-up. Seventy-seven percent of patients received additional treatment during the follow-up period, which the authors attributed to the incomplete impact of the acute treatment. Patients were found to have had additional benefit from the follow-up treatment.
Mavissakalian and Perel (1999) treated 110 patients with panic disorder for 6 months with imipramine at 2.25 mg/kg/day. Thirty patients dropped out during this period due to typical imipramine side effects. Fifty-six patients who remitted and consented were placed into a double-blind maintenance condition (n = 29) or a placebo (n = 27) and followed for 12 months. The patients were not permitted to obtain treatment interventions outside the study. Within the study, they could obtain 1 or 2 crisis intervention sessions, but not panic-focused cognitive or behavioral treatment. Relapse in the maintenance treatment (n = 1) was significantly lower than in placebo (n = 10). However, the authors also note that 8 patients dropped out of the imipramine group and 7 out of the placebo group, although these dropouts were clinically stable at their last assessment. Eleven patients required supportive sessions unrelated to panic (total of 13 visits).
In summary, then, both specific medications and cognitive-behavioral treatments have demonstrated efficacy in the short-term treatment of panic disorder. The chronic and recurrent nature of panic symptoms in many patients, however, may require the ongoing use of medication with its attendant side effects. Cognitive-behavioral treatments may aid in the reduction of vulnerability to panic recurrence, although this has yet to be clearly demonstrated. Psychodynamic treatments show promise but have not yet been subjected to efficacy studies. Current and future efforts should focus on which factors, neurophysiological and psychological, predispose to panic relapse, and to determine which interventions, or sequence of interventions, reduce vulnerability to panic recurrence.
Was this article helpful?