Reliable and feasible fitness testing for children on the autism spectrum

Purpose: This study examined the test-retest reliability and feasibility of select fitness assessments in 7–12 year old children on the autism spectrum. Method: Participants (N = 14; n = 1 female; Mage = 9.5 ± 1.7 years) completed 7 fitness assessments, administered in a random order, on two occasions: Bruce protocol; Modified 6-minute walk test (M6MWT); Wingate anaerobic cycling test; muscle power sprint test (MPST); sit & reach; standing long jump; and grip strength. Intraclass correlations (two-way mixed with absolute agreement) were computed to examine test-retest reliability. Feasibility was assessed by questionnaire following the first administration of each test. Results: The Wingate (ICC = .956), standing long jump (ICC = .925), grip strength (ICC = .913), and sit and reach (ICC = .829) tests demonstrated good- to- excellent reliability, while the Bruce protocol (ICC = .811), M6MWT (ICC = .510), and MPST (ICC = .703) demonstrated moderate- to- good reliability based on the 95% confidence intervals of the ICC. All tests demonstrated assessor-rated feasibility scores of 70/100 or higher and child-rated feasibility scores of 66/100 or higher. Conclusion: The results demonstrate moderate- to excellent test-retest reliability for select fitness tests. Short, single-instruction (e.g., standing long jump) tests may be more reliable than lengthier assessments (e.g., M6MWT) in this population. Implications of this work include the ability of practitioners and researchers to feasibly and reliably measure the fitness of school-aged children on the autism spectrum for ongoing health and behavioural monitoring and intervention purposes