Abstract/Text: Whereas lifetime prevalence rates of panic disorder--as established in epidemiologic surveys--range between 1.6 and 3.5%, 1-month rates usually amount to much less than one half of the lifetime rates. This finding indicates that a substantial proportion of patients who had panic disorder at some stage in their life must have remitted. In contrast to these results, clinicians tend to regard panic disorder as a chronic condition because, as a rule, they see panic patients only several years after onset of the disorder. A number of small, prospective, long-term studies of such clinical populations indicate that after several years, between 17 and 70% of patients still have panic attacks, and between 36 and 82% have phobic avoidance. In the largest and longest follow-up study published to date, 45% of all patients showed an unremitting--although in a certain proportion waxing and waning--course, 24% followed a pattern of remissions and relapses, whereas 31% went back into a stable remission. The evidence of factors predicting the course of panic disorder in clinical populations suggests that long duration and agoraphobia at baseline--not the severity and frequency of panic attacks--are predictors of an unfavorable course. Additional studies are needed to determine whether personality factors, depression, and other variables are also of predictive relevance. Also, factors working during follow-up, such as positive and negative life events, coping behaviors, and treatment, should be considered in future studies.