Behac. Rrs. 7-k. Vol. 20. pp. 219 to 231. 1982 COOS-7967;82/0302 19. I3SO3.00/0 Printed in Great Brttain. All nghts reserved CopyrIght 0 1981 Pergamon Press Lid zyxwvutsrqp EXPOSURE AND COMMUNICATIONS TRAINING IN THE TREA TMENT OF AGORAPHOBIA DIANNE L. CHAMBLESS, ** EDNA B. FoA,’ GERALD A. GROVES~ and ALAN J. GOLDSTEIN’ ‘Department of Psychology, University of Georgia, Athens, GA 30602; ‘Department of Psychiatry. Temple University Medical School, Philadeiphia, PA 19129; and 3Department of Psychiatry, Boston University Medical School, Boston, MA 02215, U.S.A. (Receioed 28 Augusr 1981) Summary-Twenty-one agoraphobics participated in a &month study designed to (a) compare imaginal flooding under high- (no sedation) and low- (sedation used) anxiety conditions; (b) examine the long-term effects of imaginal flooding in the absence of further exposure treatment; and (c) explore the impact of communications training on chronic anxiety and panic attacks. A reevaluation of the effects of sedation is presented in this report. Imaginal flooding without sedation was, on the whole, superior to the attention control pla- cebo and imaginal flooding with sedation on therapists’ and clients’ ratings of fear and avoidance. However, the superiority of the non-drug ffooding group cannot be attributed (as was concluded in an earlier report) to higher levels of anxiety across flooding sessions. Rather the drug fmetho- hexitone sodium) appears to have impeded across-session habituation, perhaps by interfering with long-term memorial processes. The effects of imaginal flooding without sedation were stable over 4 months without further exposure treatment. These clients did receive training in solving important interpersonal prob- lems through self monitoring and increased expressiveness. Contrary to hypotheses, not only was no further improvement obtained on fear and avoidance with this treatment, but also anxious mood and panic attacks remained unaffected. It is surprising to note that zyxwvutsrqponmlkjihgfedcbaZYXW in-uivo treatment was no more effective than imaginal flooding on fear and avoidance. These and other findings suggest imaginal flooding well warrants further study. INTRODUCTION In 1979 in this journal we reported the results of a study on the role of anxiety in flooding with agoraphobics (Chambless er al., 1979) in which we concluded that the experience of anxiety during flooding enhances the technique’s effectiveness. Anxiety in that investigation was manipulated with methohexitone sodium, a short-acting barbitu- rate. While the purpose of the current paper is to provide further data on a subsequent phase of treatment for clients in the earlier study, we have also found reason to alter our conclusions on the earlier study and have presented a reanalysis of that data as well. In previous investigations of imaginal. flooding with agoraphobics, in-vivo exposure has been provided in a second phase of treatment or during the follow-up period (Emmel- kamp and Wessels, 1975; Mathews et al., 1976; Watson and Marks, 1971). This has prevented the assessment of the longer-term effects of imaginal flooding which is particu- larly unfortunate, since the immediate outcome may well differ from the long term (e.g. Foa er al., 198Oa). One goal of this follow-up investigation was the assessment of the stability of imaginal flooding’s effects over a 4-month period during which no further exposure was conducted. Clients continued in treatment, receiving therapy directed at other problem areas: chronic, generalized anxiety and panic. In our previous report we found no change on chronic anxiety and panic with imagi- nal flooding despite changes on fear and avoidance. These findings are congruent with arguments by Zitrin et al. (1980) and Sheehan er al. (1980) who contend that these problems are endogenous in nature and are best treated with pharmacotherapy. Gold- stein and Chambless (1978), however, have proposed that these symptoms spring from poor coping with interpersonal, usually marital, conflict and can be resolved without drug treatment. A further question in the follow-up investigation, therefore, was whether * To whom all reprint requests should be addressed. 219