Development and validation of a new ataxia rating scale: Scale for the Assessment and Rating of Ataxia (SARA)

T. Schmitz-Hübsch; S. Tezenas du Montcel; L. Baliko; J. Berciano; S. Boesch; A. Durr; P. Giunti; J. Kang; C. Mariotti; M. Pandolfo; M. Rakowicz; L. Schöls; S. Szymanski; B. van de Warrenburg; T. Klockgether

doi:10.1055/s-2005-919541

Development and validation of a new ataxia rating scale: Scale for the Assessment and Rating of Ataxia (SARA)

T Schmitz-Hübsch ¹, S Tezenas du Montcel ¹, L Baliko ¹, J Berciano ¹, S Boesch ¹, A Durr ¹, P Giunti ¹, J Kang ¹, C Mariotti ¹, M Pandolfo ¹, M Rakowicz ¹, L Schöls ¹, S Szymanski ¹, B van de Warrenburg ¹, T Klockgether ¹

¹Bonn; Paris, F; Pecs, HUN; Santander, E; Innsbruck, A; London, UK; Frankfurt/Main; Milan, I; Brussels, B; Warsaw, PL; Tubingen, Bochum; Nijmegen, NL

Abstract

Validated instruments for standardized graded assessment of symptom severity are essential prerequisites for clinical trials. To date, such instruments are lacking for cerebellar ataxia. Although ataxia rating scales have been published, they have not been sufficiently validated.

We therefore developed a new scale named Scale for the Assessment and Rating of Ataxia (SARA) that allows a graded assessment of ataxia symptoms on an impairment level. Items were chosen for their specificity for cerebellar involvement and the possibility for standardized performance and rating. To determine essential biometric properties two multicentric validation trials involving 11 centers from 8 European countries were performed in 167 and 119 spinocerebellar ataxia (SCA) patients, respectively.

SARA is composed of 8 items including gait, stance, sitting, speech, finger chase, finger-nose test, fast alternating hand movements and heel-shin slide with a sum score range of 0–40 (0=no ataxia, 40=severe ataxia). The scale is supplemented by an Inventory of Non-Ataxia Symptoms (INAS) aiming at a standardized full clinical description of symptoms in SCA patients.

The first validation trial allowed to determine objectivity and practicability, as well as reliability (internal consistency, inter-rater reliability, intra-rater reliability) and validity (concurrent validity, comparison with convergent criterion). Results showed sufficient practicability with a mean performance time of 17min. Factorial analysis showed that all items except one were determined by a single factor. Internal consistency was good with Cronbach's alpha 0,90. Inter-rater reliability was very good for the sum score (0.97) but only limited (0.7) for finger-nose test. Mean sum score increased with higher ataxia disability stages. However, data were insufficient to show linearity of the scale due to bias towards less severely affected patients. Unexpectedly, side differences were reported in 33.5% of items measuring limb ataxia. According to these results, a modified scale underwent a second validation trial.

The study was performed within the European integrated project on spinocerebellar ataxias (EUROSCA), funded by the European commission. Contributing clinical centers were: Bochum (D), Bonn (D), Brussels (B), Frankfurt a.M. (D), Innsbruck (A), London (UK), Milano (I), Nijmegen (NL), Paris (F), Pecs (H), Santander (E),Tübingen (D), Warsaw (PL).