Objective: High percentages of treatment discontinuation are usually recorded among patients with panic disorder-agoraphobia (PD) treated with antidepressants. Poor adherence to treatment has been attributed to "jitteriness syndrome” and to pharmaco-phobia, especially in the first phases of the treatment, and to weight gain, sexual impairment, and anticholinergic symptoms in the long term. In the present study we examined the relationships between long-term treatment response, occurrence of side effects, and noncompliance in a large sample of PD patients treated with antidepressants.
Method: A total of 326 PD patients were naturalistically treated with antidepressants (imipramine n = 127, 39%; clomipramine n = 93, 28.5%; paroxetine n = 76, 23.3%; or "other antidepressants” n = 30, 9.2%) and followed for a period of three years. All patients were evaluated by means of the Structured Clinical Interview for Diagnosis (SCID), the Panic Disorder/Agoraphobia Interview (PDI), and the Longitudinal Interview Follow up Examination (Life-up). ANOVA and Chi-square analysis were utilized for comparative analyses. Survival analysis and logistic regression have been employed to study the relationship between treatment adherence and demographic and clinical characteristics of PD, treatment response, and side effects.
Results: During the three-year follow-up period, 179 patients (54.9%) interrupted pharmacological treatment. Forty-eight patients (26.8%) were not traceable or refused to be interviewed. Among the patients who had been interviewed, 66 (20.2%) had deemed further contact with the psychiatrist unnecessary because of PD remission. Other reasons of interruption were ineffectiveness of the treatment (n = 39, 18.4%), side effects (n = 19, 10.6%), personal reasons (n = 13, 7.3%). Patients who interrupted the pharmachological treatment because of remission of the symptomatology remained in the study for a longer period than patients who interrupted it for inefficacy. Severity of PD and agoraphobia and the length of illness were significantly lower in subjects who interrupted the treatment for remission of PD compared with the other groups, while greater symptomatological severity, and longer duration of illness predicted a better adherence to the medication regimen.
Conclusion: Our data suggest that side effects or inefficacy of the treatment do not represent the only factors of noncompliance in long-term treatment of PD with antidepressants, but also a high percentage of patients who achieved a symptomatological remission tended to default from further treatment. A more severe and long lasting symptomatology predicted a better compliance to long-term treatment with antidepressants.