Agreement between instruments for assessing balance after stroke

Abstract

Introduction – Balance disorders assessment after stroke is essential because of their frequency and consequences. The use of validated instruments, easy to apply is important for fall risk identification and allows the selection of individualized intervention strategies. Goals – Check the agreement between assessments of static and dynamic balance and functional mobility after stroke and verify if the agreement changes according to the degree of dependence. Methodology – Transversal study in a sequential sample of adults with independent walking until 2 years after stroke, referred for physiotherapy as an outpatient. The static balance was assessed by the Romberg test, the dynamic balance by the Berg Balance Scale (BBS), functional mobility by the Timed Up and Go test (TUG), and the degree of dependence by the Motor Assessment Scale (MAS). The agreement was evaluated by the inter-class correlation coefficient and the Cohen k test. Results – 52 were enrolled, 26 with mild dependence, 20 with moderate, and 6 with severe. Romberg was positive 48/52; 39/52 had BBS<45 and TUG>14 was observed in 42/52. The global agreement between the three evaluations obtained ICC = 0.63. The agreement between BBS and TUG globally obtained k = 0.61, and in subjects with mild dependence k = 0.52, and with moderate dependence k = 0.64. The agreement between Romberg and BBS globally obtained k = 0.27 and between Romberg and TUG k = 0.04. Conclusions – The agreement between the evaluations of static balance, dynamic balance, and functional mobility after stroke is virtually complete in severe dependence cases and high in subjects with moderate dependency. These results reinforce the need to use these three assessment tools, particularly in cases of mild dependence.