Accuracy of Parkinson's disease diagnosis in 610 general practice patients in the West of Scotland

Abstract

UK-based community studies have found high rates of misdiagnosis in Parkinson's disease (PD). Searches of prescription databases and case records identified 610 patients taking antiparkinson therapy for a PD diagnosis in 92 West of Scotland General Practices. Patients with no documented progression of parkinsonism and/or no increase in antiparkinson medication for 3 years were assessed by two movement disorder specialists. FP-CIT SPECT scanning was performed in clinically uncertain cases. Those considered unlikely to have PD had antiparkinson drugs tapered then stopped, with a minimum of 6 months follow-up. Age, sex and disease duration matched controls were also assessed. 64 of 89 (71.9%) patients meeting selection criteria were assessed, of whom 36 (56.3%) were appropriate for therapy withdrawal. Thirty three of those 36 patients (91.7%) and 3 of 64 (4.7%) controls stopped antiparkinson therapy without deterioration giving an overall total of 36 of 610 (5.9%). The revised diagnoses in this group were mainly essential tremor (ET) (n = 14) and vascular parkinsonism (VP) (n = 10). Patients managed in Primary Care were significantly more likely to complete therapy withdrawal than those attending a specialist clinic (15.3% vs. 2.6%, P < 0.0001). The total annual cost of antiparkinson medication for these 36 patients was 13,400 pounds; the mean duration of diagnosis was 6.8 years (SD 5.6). At least 1 in every 20 patients taking medication for PD is misdiagnosed. Nearly all of these patients can be identified by simple screening of prescription databases and case records in Primary Care, followed by clinical review, which allows withdrawal of unnecessary medication.