Abstract Importance Although the Trunk Impairment Scale (TIS) is widely used in stroke rehabilitation, the quality and certainty of evidence regarding its measurement properties have been unclear. Objective The objective was to systematically evaluate the measurement properties of the TIS in individuals with stroke, using the COSMIN (COnsensus-based Standards for the selection of health Measurement INstruments) methodology. Data Sources Five electronic databases (PubMed, EBSCOhost, CINAHL, MEDLINE, and ICHUSHI) were searched from their inception dates to January 14, 2025. Study Selection Studies assessing the psychometric properties of the TIS in individuals with stroke were included. Data Extraction and Synthesis The studies' methodological quality was evaluated using the COSMIN Risk of Bias checklist. The certainty of evidence was rated using a modified GRADE approach, and meta-analyses were conducted when appropriate. Main Outcomes and Measures The measurement properties evaluated were content validity, structural validity, cross-cultural validity, internal consistency, reliability, measurement error, criterion validity, construct validity, and responsiveness. Results Thirteen studies from 7 countries met the inclusion criteria, with sample sizes of 28 to 201 participants. Based on the modified GRADE approach, internal consistency showed "low" certainty of evidence, whereas all other measurement properties were rated as "very low" certainty due to methodological limitations, including lack of unidimensionality testing, insufficient hypothesis formulation, and poor reporting. A meta-analysis indicated acceptable pooled Cronbach α-values (mean α = .88) and high intra and interrater reliability (mean intraclass correlation coefficient ICC = 0.98), although heterogeneity and risk of bias reduced confidence in these findings. Conclusions and Relevance Despite its widespread use, the TIS lacks high-quality psychometric evidence. Further validation studies using COSMIN-aligned methodology are needed to establish its reliability and validity in diverse stroke populations.
Eguchi et al. (Thu,) studied this question.