We studied the association between preceding psychiatric disorders and Parkinson's disease (PD) using a case-control design. We used the medical records-linkage system of the Rochester Epidemiology Project to identify 196 subjects who developed PD in Olmsted County, Minnesota, during the years 1976–1995. Each case was matched by age (±1 yr) and sex to a general population control. We reviewed the complete medical records of cases and control subjects to detect preceding psychiatric disorders. The frequency of psychiatric disorders was higher in cases than in control subjects; the odds ratio was 2.2 for anxiety disorders (95% confidence interval [95% CI] = 1.4–3.4; p = 0.0003), 1.9 for depressive disorders (95% CI = 1.1–3.2; p = 0.02), and 2.4 for both anxiety disorders and depressive disorders occurring in the same individual (95% CI = 1.2–4.8; p = 0.02). When we restricted analyses to disorders present 5 years or more before the onset of motor symptoms of PD, the association with depressive disorders lost statistical significance. However, the association with anxiety disorders remained significant for disorders present 5, 10, or 20 years before onset of motor symptoms. Our results suggest that anxiety disorders and depressive disorders are associated with PD and that the causative process or the risk factors underlying PD are present many years before the appearance of motor symptoms.