Dual task interference in estimating the risk of falls and measuring change: a comparative, psychometric study of four measurements.

Objective: The aim is to examine the validity and reliability of the Timed Up and Go Test with dual task for predicting the risk of falls. Standard values for the TUG with dual task were determined, taking account of age and gender.

Design: Validation study.

Setting: Data was recorded for 120 volunteers in an outpatient physiotherapy centre.

Subjects: The sample comprised 120 healthy men and women aged 60 to 87 years living at home. Twenty-three subjects selected at random were tested again, after one day and after one week, in order to ascertain the retest reliability.

Main Measures: Berg Balance Scale, times for Timed Up and Go Test with manual dual task (TUGman) and with cognitive dual task (TUGcog).

Results: Strong correlations between the TUGman and the BBS (r = -0.72) and between the TUGcog and the BBS (r = -0.66) indicate high criterion validity. The retest reliability of the TUGman (r(T1-T2) = 0.97 and r(T1-T3) = 0.98) and TUGcog (r(T1-T2) = 0.98 and r(T1-T3) = 0.98) is very good. The intra-rater reliability is very high with an ICC = 0.99 for the TUGman and an ICC = 0.94 for the TUGcog. The mean time needed to perform the TUGman is 11.6 s (95% CI 11.2-11.9); the mean time needed to perform the TUGcog is 9.8 s (95% CI 9.5-10.2).

Conclusions: The tests with dual task can be recommended because they possess high criterion validity and very good retest reliability. Faster and simpler performance of the TUG-DT is another factor in its favour.