Assessing the efficacy of treatments for panic disorder and agoraphobia. I. Methodological problems

An assessment of the efficacy of treatments in panic disorder and agoraphobia (PDA) is complicated by methodological problems resulting from the complexity of these disorders. In this paper, commonly used anxiety scales are discussed with regard to their relevance in the assessment of PDA. Although there are already many scales for assessing anxiety and agoraphobia, a qualified instrument for measuring the severity of PDA is still lacking. A survey of the instruments used in recent treatment studies of PDA shows that rather inhomogeneous efficacy criteria were applied. Most often, panic attack frequency was taken as a severity criterion, although this measure showed little power to differentiate between treatments. To deal with the special features of PDA, a battery of different scales was usually employed and non-standardized scales were sometimes used additionally. A new rating scale designed to overcome the methodological problems associated with the assessment of severity in PDA was introduced. The use of the psychiatrist-rated Panic and Agoraphobia (P&A) scale in patients with DSM-III-R panic disorder and/or agoraphobia (n = 235) shows that this scale has the highest correlation with the psychiatrists' clinical global impression of PDA patients (r = 0.79, p < 0.0001), compared with other commonly used observer-rated anxiety scales. The self-rating version of the P&A again had the highest correlation with a patient-rated clinical global impression (r = 0.82, p < 0.0001), whereas the other patient-rated anxiety questionnaires correlated only moderately or poorly. By means of logistic regression, it was shown that the five subscores of the scale contribute independently and almost equally to the prediction of PDA severity.