Imagined Timed Up and Go test (iTUG) in people with Parkinson’s Disease: test-retest reliability and validity

Abstract

Purpose To determine the test-retest reliability and validity of the Imagined Timed Up and Go Test (iTUG) as a Motor Imagery measure of temporal accuracy in people with Parkinson’s Disease (PD).

Materials and methods A descriptive study was conducted following the GRRAS recommendations. Thirty-two people with idiopathic, mild to moderate PD (Hoehn and Yahr I–III), without cognitive impairment (MMSE ≥ 24), were assessed twice (7–15 days apart) with the iTUG. The absolute unadjusted difference in seconds, and the absolute adjusted difference as percentage of estimation error, between real and imagined TUG times, were calculated as outcome measures. Test-retest reliability was assessed using a two-way mixed-effects model of the ICC. Construct validity was tested with the Imagined Box and Blocks Test (iBBT) and convergent validity with clinical characteristics of PD, using the Spearman’s rank correlation coefficient.

Results The ICC for the unadjusted and adjusted measures of the iTUG was ICC = 0.61 and ICC = 0.55, respectively. Correlations between iTUG and iBBT were not statistically significant. The iTUG was partially correlated to clinical characteristics of PD.

Conclusions Test-retest reliability of the iTUG was moderate. Construct validity between iTUG and iBBT was poor, so caution should be taken when using them concurrently to assess imagery’s temporal accuracy.

Implications for Rehabilitation In people with Parkinson’s Disease (PD), the absolute unadjusted difference (in seconds) and the absolute adjusted difference (as a percentage of estimation error) of the Imagined Timed Up and Go test (iTUG) were moderately reliable.

iTUG and Imagined Box and Blocks Test (iBBT) measures were not statistically correlated. Therefore, temporal accuracy measures of Motor Imagery are highly task-dependant and thus their construct validity is poor.

Correlations between the adjusted and unadjusted measures of the iTUG and the majority of clinical variables of PD were not statistically significant. Statistically significant correlations were only found between the unadjusted difference and MDS-UPDRS Part III, Schwab and England, and Berg Balance scales, as well as the adjusted difference and disease duration.