Abstract Background A number of western studies have suggested that the 6-month duration requirement of generalized anxiety disorder (GAD) does not represent a critical threshold in terms of onset, course, or risk factors of the disorder. No study has examined the consequences of modifying the duration requirement across a wide range of correlates in both developed and developing countries. Methods Population surveys were carried out in 7 developing and 10 developed countries using the WHO Composite International Diagnostic Interview (total sample size = 85,052). Prevalence of GAD was estimated using different minimum duration criteria. Age of onset, symptom persistence, subsequent mental disorders, impairment, and recovery were compared across GAD subgroups defined by different duration criteria. Results Lifetime prevalence estimates for GAD lasting 1 month, 3 months, 6 months, and 12 months were 7.5%, 5.24%, 4.11%, and 2.95% for developed countries and 2.65%, 1.78%, 1.47%, and 1.17% for developing countries, respectively. There was little difference between GAD of 6 months duration and GAD of shorter durations (1–2 months, 3–5 months) in symptom severity, age of onset, persistence, impairment, or comorbidity. Those with GAD lasting 12 months or more were the most severe, chronic, and impaired of the four duration subgroups. Conclusion In both developed and developing countries, the clinical profile of GAD is similar regardless of duration. The DSM-IV 6-month duration criterion is not an optimal marker of severity, impairment, or need for early treatment. Future iterations of the DSM and ICD should consider shortening the duration requirement of GAD.