Validity and Reliability of Two Abbreviated Versions of the Gross Motor Function Measure

Abstract

BACKGROUND: The "gold standard" for measuring gross motor function in children with cerebral palsy is the 66-item Gross Motor Function Measure (GMFM-66). OBJECTIVE: The purpose of this study was to estimate the validity and reliability of 2 abbreviated versions of the GMFM-66; one version involves an item set approach, and the other version involves a basal and ceiling approach. DESIGN: This was a measurement study comprising concurrent validity, comparability, and test-retest reliability components. METHODS: The study participants were 26 children who were 2 to 6 years of age and had cerebral palsy across all Gross Motor Function Classification System levels. In the first session, both abbreviated versions were administered by 2 independent raters; next, the full GMFM-66 was administered. In the second session, only the abbreviated versions were administered by the same raters. Concurrent validity, comparability of versions, and test-retest reliability were determined with intraclass correlation coefficients [ICC (2,1)]. RESULTS: Both versions demonstrated high levels of validity, with an ICC of .99 (95% confidence interval=0.972-0.997), reflecting associations with the GMFM-66. Both versions also were shown to be highly reliable, with ICCs of greater than .98 (95% confidence interval=0.965-0.994). LIMITATIONS: A smaller-than-expected sample was recruited for this study and may be a potential limitation of the study. CONCLUSION: Both versions of the GMFM-66 can be used in clinical practice or research. However, the GMFM-66 with the basal and ceiling approach is recommended as the preferred abbreviated version.