Background: Generalized anxiety disorder (GAD) is generally considered to be a chronic condition, waxing and waning in severity; however prospective investigation of the course of GAD in community samples is lacking. This study seeks to fill that gap, by identifying the whole spectrum of generalized anxiety syndromes, sub-typing them according to their duration and frequency of occurrence, and evaluating their long-term course and outcome in the community.
Method: The prospective Zurich Study assessed psychiatric and somatic syndromes in a community sample of young adults (N = 591) (aged 20 years at first interview) by six interviews over a period of 20 years (1979-1999). GAD syndromes were defined by DSM-III symptom criteria without applying any exclusion criteria. A spectrum of generalized anxiety was defined by duration: 6 months (DSM-IV), 1 month (DSM-III), < or = 2 weeks (with weekly occurrence over one year), and anxiety symptoms. From 1978 (screening) to 1999 the annual presence of symptoms and treatment was assessed. Persistence of anxiety was defined by the almost daily presence of symptoms over the previous 12 months.
Results: The annual incidence of DSM-III GAD increased considerably between the ages of 20 and 40. The average age of onset of symptoms was 15.6 years; in 75% of cases it occurred before the age of 20. 75 of 105 DSM-III GAD cases had at least one follow-up. At their individual last follow-up, 12 of those 75 subjects (16%) were re-diagnosed as having GAD, 22 (29%) manifested subthreshold syndromes or anxiety symptoms, while 39 cases, the majority, (52%) were symptom-free; 5 of the 12 re-diagnosed GAD cases were persistent (corresponding to 7% of all 75 initial GAD cases). In their twenties they were treated at some time in 6% of all years, but in their thirties this figure rose to 12%. At their individual last follow-up 26% of 6-month GAD subjects and 22% of 1-month GAD subjects were still being treated. Treated vs. non-treated subjects did not differ in terms of gender but did differ in severity, persistence and in comorbidity with bipolar-II disorder, social phobia, obsessive-compulsive syndromes and substance-use disorders.
Limitations: Results are based on a relatively small sample and cannot be generalized to adults aged over 40 years.
Conclusions: The course of DSM-III-defined GAD may not be chronic, as previously suggested, but mainly recurrent with intervening symptom-free periods of recovery in about half of cases. Over a period of 20 years there was more improvement than progression within the anxiety spectrum.