Correlations of Balance & Gait Measures with the UPDRS and with Each Other


Abstract The purpose of the study was to determine whether a)current balance and gait measures correlate with the Unified Parkinson’s Disease Rating Scale (UPDRS) total and motor scores; b)performance and self-perceived measures of balance correlate; c)instrumented and non-instrumented over ground measures of gait variables are related. We recruited 89 subjects with mild to moderate Parkinson’s disease (PD), 65 male, mean age 69.73 ± 8.18 years, mean time since onset 8.25 ± 5.23 years, to perform balance and gait tests. Patients walked forward at usual and fast speeds and backwards on the instrumented walkway, performed timed tests of balance and walking over ground, as well as the Gait and Balance Scale (GABS), and the UPDRS while off and on PD medication. Subjects also rated balance confidence. Spearman’s rho correlations were calculated for each balance and gait parameter and the UPDRS, for performance and self-perceived balance measures, and for gait parameters walking on the instrumented walkway and over ground. Paired t-tests were performed medication to compare means for the gait variables on the instrumented walkway and over ground. We found that the postural instability gait dysfunction index (PIGD) was the only balance or gait measure to correlate with the UPDRS total score exclusively off PD medications. None of the other balance or gait measures were significantly associated with the total or motor section scores of the UPDRS. The performance balance measures GABS were strongly related to the perceived and each other. The strongest correlations between instrumented and non-instrumented walking were for velocity walking fast forward, both off and on medications and for stride length walking backward both off and on medication. Means for velocity walking forward and fast forward, cadence walking forward, and stride length walking backward were significantly higher on the instrumented walkway than over ground without instrumentation. Cadence walking backward was higher over ground off and on medication. We concluded that dynamic tests of balance and walking probably are measuring constructs different than the UPDRS. While performance and self-perceived tests of balance correlate strongly, it is probably necessary to measure both. Instrumented and over ground walking need further study.



Parkinson's disease, evaluation, balance, gait, UPDRS