Evaluating the Test-Retest Reliability of Five Low-Cost, Perturbation-Based Functional Tests for Balance Recovery in Older Adults.

BACKGROUND: Falls are a leading cause of injury and disability among older adults. Conventional clinical tests typically do not challenge reactive postural responses to unexpected perturbations, which limits their ability to comprehensively assess fall risk. OBJECTIVE: To examine the test-retest reliability of five pragmatic, low-cost, perturbation-based tests designed to identify compensatory stepping strategies in older adults, and to explore their concurrent validity against established clinical assessments. METHODS: Fifty-seven older adults (44 community-dwelling and 13 institutionalized) completed five compensatory stepping tests (obstacle crossing, forward push, backward pull, and lateral pulls to the right and left) and conventional functional tests [Timed Up and Go (TUG), 30 s Chair Stand, and the Short Physical Performance Battery (SPPB)] on two separate days, ten days apart. Cohen's weighted kappa (Kw) quantified test-retest reliability, and Pearson's correlation coefficients assessed relationships with conventional tests. RESULTS: Obstacle (Kw = 0.443), forward push (Kw = 0.518), and backward pull (Kw = 0.438) demonstrated moderate agreement overall. Lateral pull tests showed poor reliability. Nevertheless, moderate correlations were observed between some perturbation tests (particularly obstacle and backward pull) and standard clinical measures, especially TUG and SPPB. CONCLUSIONS: Although reliability was limited-most notably for lateral perturbations-specific tests showed meaningful associations with validated functional assessments. Pending methodological refinements, these low-cost tools may offer useful insights for initial fall-risk screening.