Aim: Many studies have been performed upon the methodological qualities of the (modified) Ashworth Scale, but overall these studies seem insufficiently conclusive. Aim of this study is to investigate the construct validity and inter-rater reliability of the Ashworth Scale (AS) for the assessment of spasticity in upper and lower extremities.
Method: A cross-sectional study on spasticity in the elbow flexors (part 1) and knee extensors (part 2) was carried out. In both parts AS was assessed, while muscle activity and resistance was recorded simultaneously, in patients with upper motor neuron syndrome. Each patient was measured by three raters.
Results: Thirty patients participated, nineteen in each part of the study. For elbow flexor muscles, AS was not significantly associated with electromyographic parameters, except for rater 2 (rho = 0.66, p < 0.01). A moderate significant association was found with resistance (0.54 ≤ rho ≤ 0.61, p < 0.05). For knee extensors, AS scores were moderately associated with muscle activity (0.56 ≤ rho ≤ 0.66, p < 0.05) and also with resistance (0.55 ≤ rho ¡Ü 0.87, p < 0.05).
Intraclass correlation coefficient for absolute agreement was 0.58 for elbow flexors and 0.63 for knee extensors. In linear mixed model analysis the factor Rater appeared to be highly associated with AS.
Conclusion: Validity and reliability of the Ashworth Scale is insufficient to be used as a measure for spasticity.