INTERRATER AND INTRARATER RELIABILITY OF THE MODIFIED ASHWORTH SCALE FOR CHILDREN. Nancy A. Clopton*; Robin Broom; Krystal Delgado; Jennifer Dutton; Tandy Featherston; Annette Grigsby; Melinda Hughs; Jami Mobley; Katherine Powers Rehabilitation Sciences, Texas Tech University Health Sciences Center, Lubbock, TX PURPOSE: To investigate interrater and intrarater reliability of the Modified Ashworth Scale to assess spasticity in children. BACKGROUNDS/SIGNIFICANCE: The Modified Ashworth Scale (MAS) is a 6-point scale commonly used to assess interventions for spasticity such as medication, injection of botulinum toxin, selective posterior rhizotomy, or implantation of an intrathecal baclofen pump. All of these procedures are widely used as interventions for children who have spasticity. SUBJECTS: A pilot study included 15 children. The current study included 17 children. All children in both studies had spasticity and were between 2 and 17 years of age. METHODS AND MATERIALS: Muscles tested were elbow flexors, hip adductors, quadriceps, hamstrings, gastrocnemius and soleus. Only muscles in affected extremities were tested (e.g. both lower extremities for a child with diplegia or one side only in hemiplegia). Two or three raters tested each muscle in each session. Testing was repeated one week later. In the pilot study, training for the raters was brief, reflecting the manner in which the MAS is typically used. In the current study, training was more extensive. In addition in the current study, when one examiner performed the test, an observer also recorded a score. The rationale was that a correlation of the observational and examiner that was higher than the intra-rater retest correlation would suggest that actual changes in spasticity might be artificially depressing measured reliability. ANALYSES: The Intraclass Correlation Coefficient (ICC) was computed to assess reliability and to control for chance agreement. Fleiss suggests that the ICC is the mathematical equivalent of the weighted kappa for ordinal data, but it can also assess reliability for more than two raters at a time and for different numbers of raters for each subject. The ICC can be used for ordinal data with equal distance between intervals. Based on the critique of Pandyan et al., MAS scores were considered ordinal and a value of 1.5 was assigned to ratings of 1+ to maintain equal intervals. Scores from right and left sides of the body were combined for the same muscle. RESULTS: Only significant correlations are reported. In the pilot study, ICC values for interrater reliability were moderate (ICC = .50 to .75) for elbow flexors and quadriceps, and poor (ICC < .50) for gastrocnemius, soleus, hip adductors, and hamstrings. Only the elbow flexors and quadriceps were found to have a statistically significant relationship for intrarater reliability. Reliability for quadriceps was good (ICC > .75) and moderate for elbow flexors. In the current study, interrater reliability correlations were good for elbow flexors and hamstrings and poor (ICC < .50) for other muscles. Intrarater scores were good (ICC > .75) for hamstrings and moderate (ICC = .50 to .75) for other muscles. Correlations of observational scores with examiners’ scores were good for hamstrings, poor for gastrocnemius and moderate for other muscles. CONCLUSIONS: Reliability of the MAS may be lower than desired for clinical use for muscles other than hamstrings. Training improved reliability slightly, but reliability for rating most muscles was still not in the acceptable range. Results of the attempt to assess actual variability of spasticity were inconclusive, but raters agreed that actual variability in spasticity for children appeared subjectively to be a significant factor affecting measured reliability of the MAS. FUNDING SOURCE: The study was funded by the Master in Physical Therapy Program of Texas Tech University Health Sciences Center. KEYWORDS: Cerebral Palsy, Reliability, Children, Spasticity, Muscle Tone Copyright 2009 by the American Physical Therapy Association. Requests for reprints should be directed to the corresponding author of the article. Educators, students, and other academic customers may receive permission to reprint copyrighted material from Physical Therapy (ISSN 1538-6724) by contacting the Copyright Clearance Center Inc, 222 Rosewood Dr, Danvers, MA 01923. Other types of customers who want permission to reprint should contact the APTA Editorial Office, Attn: Physical Therapy.