Clinical associations during the year before Parkinson's disease diagnosis in Taiwan: a nationwide claims-based case-control study.

Abstract

Background Characterizing clinical associations before Parkinson's disease (PD) diagnosis may improve understanding of the diagnostic window in real-world care. Using nationwide claims data from Taiwan, clinical patterns and medication exposures recorded before incident PD were examined, and secondary neurological outcomes and all-cause mortality were evaluated during follow-up. Methods A retrospective, population-based case-control study was conducted using Taiwan's National Health Insurance Research Database (2000-2016). Incident PD cases were matched 1:4 to controls using a two-stage approach: frequency matching on birth year and sex, followed by propensity score matching for healthcare utilization and selected comorbidities. Associations between prespecified comorbidities, medication exposures, and incident PD were estimated using multivariable logistic regression, while Kaplan-Meier methods and Cox proportional hazards models assessed secondary neurological outcomes and all-cause mortality. Results A total of 2124 patients with PD and 8496 matched controls were included. Within the one-year pre-diagnostic window, incident PD was significantly associated with head injury (adjusted odds ratio [aOR] 2.16), ischemic stroke (aOR 2.90), hepatitis C virus infection (aOR 1.40), hospitalization for bacterial infection (aOR 1.43), chronic exposure to propranolol (aOR 1.72), and proton pump inhibitors (aOR 1.24), whereas hyperlipidemia showed an inverse association (aOR 0.67). During follow-up, PD patients had higher adjusted hazards for secondary neurological outcomes (adjusted hazard ratio [aHR] 2.24; 95% confidence interval [CI] 1.94-2.58) and all-cause mortality (aHR 1.72; 95% CI 1.57-1.89) than controls. Conclusions These findings highlight priority variables for understanding the complex clinical landscape preceding a formal PD diagnosis.