The effect of the testing environment on the reliability of the movement assessment battery for children -2

Background: To be considered ecologically valid, a test should reflect tasks performed in the 'real' world (Campbell et al. 2009). Physiotherapists work in many different settings, but whether the results of the MABC-2 (Henderson et al. 2007) are transferrable between these settings is unknown, as the effect of the testing environment on performance has yet to be investigated. Objectives: To determine whether the MABC-2 is a reliable means of assessing motor skill proficiency in a population of 7-10 year old primary school children when applied in 2 different locations on separate occasions. Methods: 11 2nd class children (11 male, mean age 7.55yrs SD 0.52yrs) were tested on 2 occasions, 2 weeks apart, using the MABC-2. The 1st testing session took place in the children's school, the 2nd in a physiotherapy clinic, a setting unfamiliar to the participants. Results: The total test score of the MABC-2 (ICC 0.91, 95% CI 0.70 – 0.97) was shown to have good test-retest reliability (Portney and Watkins 2000). The test components, manual dexterity (ICC 0.81), aiming and catching (ICC 0.76) and balance (ICC 0.79) were shown to have good reliability, but many of the individual components were shown to have moderate (ICC 0.50-0.74) or poor reliability (ICC<0.50). Conclusions: The total test score of the 2nd age band of the MABC-2 appears to have good test-retest reliability between familiar and unfamiliar locations. Further research needs to be done to determine whether location affects motor performance in children using a larger sample size.