Antidepressants used most often in Parkinson’s disease are potentially interfering with the disease. The association between the start of antidepressants with both Parkinson’s disease and its course was therefore analysed in follow-up studies. Datawere obtained from the PHARMO database, which includes information on drug dispensing for all 300 000 residents of six Dutch cities between 1991 and 1998. All patients aged 55 years and older who used levodopa for at least 180 days were selected, and matched to at most three controls for age, gender, pharmacy, and calendar-time. Poisson regression analysis was used to study the association between Parkinson’s disease and start of antidepressants. The association with the course of disease was studied, using the Cox proportional hazards model, among patients who started an antiparkinsonian drug in the database. Identified were 633 patients with Parkinson’s disease and 1598 matched controls. During follow-up, 96 patients with Parkinson’s disease and 140controls started an antidepressant; relative risk was 1.51 (95% confidence interval 1.17-1.96). A tricyclic antidepressant was started in 65 patients and 74 controls; relative risk was 1.94 (1.39-2.71). A selective serotonin reuptake inhibitor (SSRI) wasstarted in 22 patients and 53 controls; relative risk was 0.92 (0.56-1.51). Among the 295 patients who had started an antiparkinsonian drug and their 691 controls, 23 patients and 15 controls used an antidepressant at start of follow-up; relative risk was 3.59 (1.90-6.78). During the first 6 years of follow-up, 26 patients with Parkinson’s disease and 45 controls initiated treatment with an antidepressant; relative risk was 1.18 (0.73-1.92). Patients with Parkinson’s disease start antidepressants morefrequently. At start of antiparkinsonian drug treatment, patients use antidepressants three times more often than controls. During the first 6 years of treatment with antiparkinsonian drugs, patients do not start antidepressants more frequently. Physicians who prescribe antidepressants to patients with Parkinson’s disease seem to have a preference for tricyclic antidepressants.