Investigating the internal validity of the Trunk Impairment Scale (TIS) using Rasch analysis: the TIS 2.0

Purpose. To examine the internal validity of the static sitting balance, dynamic sitting balance, and coordination subscales of the Trunk Impairment Scale (TIS), a reliable and valid scale measuring trunk performance and sitting balance in people after stroke. Method. A total of 162 people after stroke were included in the study. Participants were recruited from an acute unit and in- and out- patient rehabilitation setting. To examine internal validity of the subscales of the TIS, we conducted a Rasch analysis by means of the Partial Credit Model. For each subscale, we examined whether the distribution of scores fitted the theoretical Rasch model. Results. The first item of the static sitting balance subscale had to be removed since it had a large ceiling effect. The remaining static sitting balance subscale did not fit the Rasch model (Chi-square = 7.03, p < 0.0001 with Bonferroni adjusted p-level = 0.01). Both the dynamic sitting balance (Chi-square = 42.65, p = 0.0052 with Bonferroni adjusted p-level = 0.005) and coordination subscales (Chi-square = 7.87, p = 0.4461 with Bonferroni adjusted p-level = 0.01) fitted the Rasch model. Conclusions. Internal validity of the dynamic sitting balance and coordination subscales was confirmed. Based on our results, we present the TIS, version 2.0 (TIS 2.0).