Background. This study questions the 6-month duration criterion for generalized anxiety disorder (GAD) used in DSM-III-R and DSM-IV.

Method. In adults from age 20/21 to 40/41 in the prospective Zurich Cohort Study, four groups of generalized anxiety syndromes defined by varying duration (2 weeks, 1 month, 3 months and 6 months) were compared.

Results. Applying DSM-III (1979–1999) and DSM-III-R (1986–1999) criteria, there were no significant differences between the four groups in terms of family history of anxiety, work impairment, distress, treatment rates or co-morbidity with major depressive episodes (MDEs), bipolar disorder or suicide attempts. Only social impairment related to the length of episodes. The 6-month criterion of DSM-III-R and DSM-IV GAD would preclude this diagnosis in about half of the subjects treated for generalized anxiety syndromes.

Conclusions. In this epidemiological sample, the 6-month duration criterion for GAD could not be confirmed as clinically meaningful. GAD syndromes of varying duration form a continuum with comparable clinical relevance.