Reliability of timed walking tests and temporo-spatial gait parameters in youths with neurological gait disorders

BACKGROUND The 10-Meter Walk Tests (10MWT) and the 6-Minute Walk Test (6MinWT) are applied to assess gait capacity in paediatric patients. To better objectify changes in qualitative aspects of gait, temporo-spatial parameters like stride length or step symmetry could be simultaneously assessed with a GAITRite system. Reliability has not yet been evaluated in a heterogeneous sample of children with various neurological gait disorders such as is representative for paediatric neuro-rehabilitation. The aim of this study was to assess test-retest reliability of the 10MWT, the 6MinWT and simultaneously recorded gait parameters captured with the GAITRite system in children with neurological gait disorders. METHODS This is a cross-sectional study with two measurement time-points. Thirty participants (9 females; mean (standard deviation) age 13.0 (3.6) years, 10 with cerebral palsy, 6 after stroke, among other diagnoses) performed the 10MWT at preferred (10MWTpref) and maximum speed (10MWTmax) and the 6MinWT on two occasions (mean time interval: 7.0 (1.9) days). Relative reliability was quantified with an intra-class correlation coefficient (ICC); the measurement error reflecting absolute reliability was quantified with the standard error of measurement and the smallest real difference. RESULTS ICCs of timed walking tests (time measured with a stopwatch, step count for the 10MWT and walking distance for the 6MinWT) ranged from 0.89-0.97. ICCs of temporo-spatial gait parameters ranged from 0.81-0.95 (10MWTpref), from 0.61-0.90 (10MWTmax) and from 0.88-0.97 (6MinWT). In general, absolute reliability was greatest in the 6MinWT. CONCLUSION Timed walking tests and temporo-spatial gait parameters obtained from the GAITRite system appear reliable in children with neurological gait disorders. However, especially in children with poorer walking ability, the reliability of temporo-spatial parameters might have been positively influenced, as unclear steps had to be removed using the GAITRite software. As absolute reliability is rather low, the responsiveness of these measures needs to be further evaluated.