Use of antidepressants and the risk of Parkinson's disease in the Local Health Trust of Bologna: A historical cohort study

Background: Depression is considered one of the prodromal symptoms of Parkinson's disease (PD) along with sleep disorders, hyposmia and constipation. Prodromal symptoms refer to the stage wherein early motor symptoms and signs allowing a diagnosis of PD are not yet present. The objective of this study was to investigate the association between the use of antidepressants, as indirect measure of depression, and subsequent PD onset, clinically diagnosed, in the Local Health Trust of Bologna, Italy. Methods: Historical cohort study with use of antidepressants as exposure and PD onset as outcome. The cohort considered consisted of inhabitants of Bologna aged >= 35 years in 2005; those who had used antidepressants in the previous 3 years were excluded. Subjects were followed up from 2006 and until PD onset, migration out of Bologna, death or end of the study period (2017), whichever came first. The ParkLink Bologna system was used to detect disease onset. ParkLink Bologna is a research study including patients with a clinical diagnosis of PD residing in Bologna. Residents that used antidepressants for at least 180 consecutive days within 1 year were considered exposed. Hazard ratios (HR) and 95% confidence interval (CI) were estimated with Cox proportional hazards models, using exposure as time-dependent variable and adjusting for potential confounders: age, gender, use of medical care and comorbidities. Results: From 2006 to 2017 199,093 person-years were exposed and 4,286,470 not exposed. Fifty-one subjects with PD were identified in the exposed group and 556 subjects in the non-exposed showing an association of adjusted HR = 1.7 (CI 1.3-2.3). The association was stronger for males (HR 2.2, CI 1.5-3.2) compared to females (HR 1.2, CI 0.8-1.9), for subjects <= 65 years of age (HR 2.4, CI 1.6-3.6) vs. > 65 years (HR 1.3, CI 0.8-1.9) and for those with less comorbidities. Age and gender were confounders in the associations between antidepressant use and PD onset. Conclusions: The use of antidepressants as indirect measure of depression is associated with the subsequent development of PD. Our findings confirm that depression may precede the onset of motor symptoms in PD. The association is stronger for younger subjects, who are males and with fewer comorbidities.