Dianosing Parkinson's disease using videotaped neurological examinations: Validity and factors that contribute to incorrect diagnoses

Field work is commonly required in movement disorders research. Sending neurologists into the field can be logistically challenging and costly. Alternatively, neurological examinations may be videotaped and reviewed later. There is little knowledge of the validity of the videotaped neurological examination in the diagnosis of Parkinson's disease (PD). We examined the validity of the videotaped Unified Parkinson's Disease Rating Scale (UPDRS) motor examination in the diagnosis of PD, and sought to determine which factors are associated with incorrect diagnoses. PD patients and controls were enrolled in a familial aggregation study between August of 1998 and June of 2000, and as part of that study each was examined by a physician who performed an in-person UPDRS motor examination. Each also underwent a second, videotaped UPDRS motor examination. Based on the review of this videotape, a neurologist, who was blinded to the previous clinical diagnosis, assigned a diagnosis of PD or normal. A total of 211 of 231 PD patients (sensitivity = 91.3%), and 170 of 172 controls (specificity = 98.8%) were correctly identified based on the videotape. True positives had a higher mean rest tremor score (1.7 vs. 0.3; P < 0.001), action tremor score (0.9 vs. 0.3; P < 0.001), bradykinesia score (11.2 vs. 7.4; P = 0.02), and disease of longer mean duration (8.9 vs. 5.8 years; P = 0.001) than false negatives. False negatives did not differ from true positives in terms of age, total dose of levodopa, Hoehn and Yahr score, or rigidity, gait and posture, or facial masking scores (each assessed during the in-person examination). The videotaped UPDRS motor examination is a useful means of diagnosing PD and provides an alternative approach for the diagnosis of PD in field studies. A limitation is that patients with milder PD of shorter duration may not be recognized as PD.