Validity and reliability of two abbreviated versions of the gross motor function measure

Validity and reliability of two abbreviated versions of the gross motor function measure
1 The University of Western Ontario, London, ON;
2 The School of Physical Therapy. The University of Western Ontario, London, ON;
3McMaster University, Hamilton, ON, Canada

Background/Objectives: The current “gold standard” measure to obtain an estimate of gross motor functioning in children with cerebral palsy (CP) is the 66-item gross motor function measure (GMFM-66). Two abbreviated versions have been developed based on Rasch analysis. One is the GMFM-66-B&C (basal and ceiling approach, in which a minimum of 15 items are administered, based on a basal of three consecutive ‘3s’ and a ceiling of three consecutive ‘0s’). The objectives of this study were to: (1) examine the concurrent validity of the abbreviated versions with the GMFM-66; (2) examine the test-retest reliability of both abbreviated versions and (3) compare the scores obtained and the time to completion of the two versions. We hypothesized that both measures would be reliable and valid, but that one would require less time to administer.
Design: This was a measurement study comprising validity, reliability and efficiency of two abbreviated versions of the GMFM-66.
Participants and Setting: The sample comprised 26 (18 males, 8 females) children with CP, between the age of 2 to 6 years, across all GMFCS levels. The children were recruited from 4 rehabilitation centres in South-western Ontario. Parents provided written consent and all study sessions took place within the centres.
Materials/Methods: Data was collected by 10 trained and reliable raters. The children took part in 2 testing sessions. The first comprised administration of both abbreviated versions of the GMFM-66 (version and order of testing randomly allocated to two raters). One consistent rater then administered the full GMFM-66. The second session (2 weeks later) comprised administration of only the two abbreviated versions, repeated in the same order and by the same rater as at the first session. GMFM-66 scores were obtained using the GMAE software. Concurrent validity and test-retest reliability were analyzed using the intraclass correlation coefficient (ICC 2.1). Comparability of the scores and time to completion were analyzed by a paired t-test. All statistical analyses were performed using SPSS 17.0.
Results: Concurrent Validity was high, GMFM-66-IS ICC=0.994 (0.987-0.997) GMFM-66-B&C ICC=0.987 (0.972-0.994) Test-retest Reliability was also high GMFM-66-IS ICC=0.986 (0.969-0.994) GMFM-66-B&C ICC=0.994 (0.987-0.997) Time to Completion is expressed as Mean (SD) GMFM-66-IS Time 1=29.16 (13.85) Time 2=21.08 (7.78) There were no significant differences in scores between the two abbreviated versions at either session. The time to completion was significantly different at both time points with the GMFM-66-B&C requiring less time to administer (P<0.01).

Conclusions/Significance: Both abbreviated versions are valid and reliable; the GMFM-66-B&C requires less time to complete. Either abbreviated version is appropriate for use in clinical practice and research to enhance efficiency of obtaining an estimate of gross motor functioning of children with CP.

Gal 1 Gal 2
Supporting Organizations
Online Feature Partner
Locations of visitors to this page

© 2010 all rights reserved. maintained by Classic Infomedia