In Reply.—
Dr Grunhaus was mis- taken in assuming that the 0 to 8-point panic scale we used corresponded to the literal number of panic attacks per week. In fact the anchoring points for the scale concerned the number of panic attacks in the past week, as follows: 0 indicated no panic attacks; 2, at least one; 4, a few times; 6, once a day; and 8, several times a day. A score of 1.5 during weeks 14 to 52 thus denoted less than 1 panic attack a week. The patients had their agoraphobia for a mean of eight years before treatment. The spontaneous cure rate in such patients is very low and unlikely to account for the favorable results in the placebo (or the imipramine) group1 (I.M.M., H. Water, MBChB, P. Lindley, RMN, unpublished data, 1983). Improvement in both patient groups continued beyond the 52 weeks reported in our article,