BACKGROUND There are concerns that the prevalence of mental disorder is increasing. AIMS To determine whether the prevalence of common adult mental disorders has increased over time, using age-period-cohort analysis. METHOD The study consisted of a pseudocohort analysis of a sequence of three cross-sectional surveys of the English household population. The main outcome was common mental disorder, indicated by a score of 12 or above on the Revised Clinical Interview Schedule (CIS-R). Secondary outcomes were neurotic symptoms likely to require treatment, indicated by a CIS-R score of 18 or over, and individual subscale scores for fatigue, sleep problems, irritability and worry. RESULTS There were 8670 participants in the 1993 survey, 6977 in the 2000 survey and 6815 in the 2007 survey. In men a significant increase in common mental disorder occurred between the cohort born in 1943-9 and that born in 1950-6 (odds ratio 1.4, 95% CI 1.1-1.9) but prevalence in subsequent cohorts remained largely stable. More extended increases in prevalence of sleep problems and mental disorders were observed in women, but not consistently across cohorts or measures. CONCLUSIONS We found little evidence that the prevalence of common mental disorder is increasing.