Abstract
BACKGROUND: The American Academy of Neurology Parkinson Disease (PD) quality measures include an annual diagnostic review. OBJECTIVE: To investigate the frequency and pattern of changes in diagnoses between PD and other causes of parkinsonism. METHODS: This prospective longitudinal cohort study included consented patients diagnosed with PD at least once and a minimum of two times at the Movement Disorders Center between 2002 and 2017. Movement disorder specialists confirmed and documented diagnoses at every visit. Longitudinal changes in diagnoses were identified across visits. RESULTS: Of 1567 patients with parkinsonism, 174 had non-PD parkinsonism with no change over time. Of 1393 patients diagnosed with PD at least once, 94% (N = 1308) had no change of diagnosis over time and 6% (N = 85) had a change of diagnosis including PD ⟷ drug-induced parkinsonism (DIP) (27.1%), PD ⟷ multiple system atrophy (MSA) (20.0%), PD ⟷ progressive supranuclear palsy (PSP) (18.8%), PD ⟷ Lewy body dementia (DLB) (16.5%), PD⟷ vascular parkinsonism (9.4%), more than two diagnoses (4.7%), and PD ⟷ corticobasal syndrome (CBS) (3.5%). The direction of diagnostic switches was as follows: PD ⟶ other parkinsonism diseases (36.5%), other parkinsonism diseases ⟶ PD (31.8%), and 31.8% of multiple switches. There were no significant differences in duration of follow-up, age at first visit, gender, race, marital status, education, income, cognition, or employment between the stable and unstable groups. Diagnostic change was associated with greater PD severity and greater medical comorbidity. CONCLUSION: Over a 15-year period, movement disorder specialists changed their clinical diagnosis of PD in 6% of patients. The most common diagnostic switches, to or from PD, were DIP, MSA, PSP, and DLB. This study describes routine clinical diagnostic patterns in the absence of pathologic confirmation. The presence of diverse diagnostic changes over time underscores the value of confirming PD diagnosis.