Changes in diagnosis with follow-up in an incident cohort of patients with parkinsonism

Background: Accurate diagnosis of the cause of parkinsonism during life can be difficult, particularly at presentation, but few studies have described changes in clinical diagnosis over time and the effect of applying strict research criteria. Methods: Incident patients with a possible/probable diagnosis of degenerative or vascular parkinsonism had a standardised assessment at diagnosis and at yearly intervals thereafter at which the most likely clinical diagnosis was recorded without strict application of research criteria. Four years after the beginning of the incident period, formal research criteria were applied retrospectively using patient records at baseline and the latest yearly follow-up. Results: Of 82 incident patients, 66 underwent at least 1 year of follow-up. After a median follow-up of 29 months, clinical diagnosis had changed in 22 (33%). Most (82%) changes occurred in the first year and were due to the development of atypical clinical features, particularly early cognitive impairment; the results of brain imaging; responsiveness to levodopa; and the rate of disease progression. Diagnosis on research criteria differed from latest clinical diagnosis in eight participants (12%). Research criteria gave a "probable'' diagnosis in 71% of parkinsonian patients at follow-up but in only 15% at the initial assessment. Discussion: The clinical diagnosis of the cause of parkinsonism at presentation was often incorrect, even when made by those with a special interest. In particular, Parkinson's disease was overdiagnosed. Research criteria were often unhelpful in clarifying the diagnosis, even after a median of 29 months of follow-up. Further research is required to identify factors that may be used to improve the accuracy of diagnosis at initial assessment.