Objectives: This study aimed to identify elderly fallers and non-fallers by balance test under dual tasks conditions.

Methods & Materials: This study was an analyze-comparative study. Subjects were from three park of Tehran. Subjects were 20 older adults with outhistory of falls (aged 75.95±6.28 years) and 21 older adults with a history of 2 or more falls in the previous one year (aged 72.50±7.31 Years) . All subjects performed Timed Up & Go test under 3 conditions (TimedUp & Go, Timed Up & Go with numbers counter randomly [TUG cognitive], and Timed Up & Go while carrying a full cup of water [TUG manual]). A multivariate analysis of variance and logistic regression analyses were performed.

Results: The results showed significant difference between elderly fallers and non fallers in fall risk composed dependent variable (P=0.0005), as the non fallers had greater score than the elderly fallers. Also, results showed that TUG cognitive has prediction capacity of elderly fall (P=0.013).

Conclusion: Consequently, balance under cognitive dual task conditions could be useful method in identification of risk of falling and planning dual task exercise program and physiotherapy to preventfalls.