Reliability and validity of the Trunk Impairment Scale in children and adolescents with cerebral palsy.

Standardized clinical tools are useful for treatment planning and evaluation, however clinical tools to assess quality in trunk movements in children with cerebral palsy (CP) are sparse. We have recently reported good intra- and inter-observer reliability of the Trunk Impairment Scale (TIS) in 5-12 year old children with CP. The aim of this study was to assess reliability in adolescents (13-19 years old), and to assess the construct validity in children and adolescents in the whole age spectrum from 5 to 19 years. Video recordings of 17 children with CP with Gross Motor Function Classification (GMFCS) level I-IV were analyzed by three observers on two occasions. For construct validity the TIS was compared with Gross Motor Function Measure (GMFM), in 37 children with GMFCS levels I-IV. Intraclass correlation coefficients varied between 0.82 and 0.98, and 86% of the kappa values varied between 0.61 and 1.00, suggesting high inter- and intra-observer reliability. The smallest detectable difference (SDD) of the TIS (scale range 0-23) varied between 2.55 and 3.82 for intra- and 4.07-8.23 for inter-observer observations. The high inter-observer SDD was partly due to consistently lower TIS scores by one observer. The correlation between the TIS total score and the dimension scores of the GMFM was high (Spearman's rho: 0.80-0.87), while decreasing GMFCS levels were associated with increasing total TIS score; both findings indicating good construct validity of the TIS. This study suggests that the TIS is a reliable and valid measure of trunk control for both children and adolescents with cerebral palsy.