Reliability of clinical diagnosis of dystonia

Abstract

Background: There is only one small single-center study on the reliability of the diagnosis of focal dystonia. The aim of this study was to assess the inter-rater reliability of dystonia diagnosis among neurologists with different professional experience.

Methods: Twenty-nine adults (18 with dystonia, 9 with other movement disorders, and 2 healthy controls) were videotaped while undergoing neurological examination and during the process of collecting information on the history of their condition. Each case was diagnosed by 35 blind raters (12 general neurologists, 21 neurology residents, and 2 experts in movement disorders) from different hospitals. Sensitivity and specificity were calculated confronting raters with the gold standard (the caring physician). Inter-rater agreement was measured by the Kappa statistic.

Results: Specificity and sensitivity were 95.2 and 66.7%, 76.3 and 75.2%, 84.6 and 71.6% for experts, general neurologists, and residents, respectively. Kappa values on dystonia diagnosis ranged from 0.30 to 0.46. The agreement was moderate for experts and residents (0.40-0.60) and fair for general neurologists (0.20-0.40). Kappas were the highest among experts for cranial and laryngeal dystonia (0.61-1), but not for cervical dystonia (0.37).

Conclusions: The diagnosis of dystonia is difficult and only partially mirrors a physician's background.