Audible tremor: Analysis of the validity of audible information in assessing tremor severity

Assessing the severity of kinetic tremor is important in clinical and research settings. Archimedes spirals are often used to assess tremor severity. Rating tremor from spirals has been based solely on visual information. However, one can often "hear" tremor.

We evaluated the type of information that could be gained from hearing tremors and assessed its validity using visualized tremor as a gold standard.

52 essential tremor cases (94 spirals), 11 Parkinson's disease patients and 11 controls (20 spirals each) were enrolled. During videotaping, cases drew Archimedes spirals in a standardized manner. Three experienced movement disorders neurologists first listened to videotaped spiral drawing (no visual) and assigned a tremor severity rating using Washington Heights Inwood Genetic Study of Essential Tremor (WHIGET) ratings (0, 0.5, 1, 1.5, 2, 3). Later, the neurologist rated the physical (i.e., visual) spiral and assigned a WHIGET tremor rating.

The Spearman's correlation coefficient (visual vs. auditory rating) was 0.721 (p < 0.001) and the weighted kappa statistic was 0.674, a level of agreement that is considered good.

Auditory information provides in-time synchronous corroboration of what the rater is seeing. That is, it informs the visual data and provides construct validity. The results of this study have ramifications for observational and experimental therapeutic studies, which commonly rely on videotaped tremor examinations. Additional corroborative information is available if the sound is left on. Regardless of the decision to present sound to the rater or not, clear instructions and a standardized practice within the study are imperative.