INTER-RATER RELIABILTY IN THREE RAPID STEPPING TESTS CHALLENGING MEDIAL-LATERAL WEIGHT-SHIFTING MOTION

S. Hannigan-Luther, B.R. Lazzarini Willamette University, Salem, OR The control of medial-lateral (ML) motion during standing balance and locomotion becomes increasingly difficult with aging. This aspect of function can be an early indicator of fall risk in older adults, but the ability to measure this in the clinic is limited. This analysis is part of a project to develop a clinically-feasible test of rapid stepping performance that challenges medial-lateral motion in older adults. PURPOSE: To determine the inter-rater reliability for three tests challenging rapid stepping performance and ML control of weight-shifting motion. METHODS: Older adults aged 70-96 yrs performed three rapid stepping tests: 1) the Step in Place test (SIP, n = 43) where participants stepped in place as fast as safely possible until reaching 20 footfalls of the first stepping foot, 2) the Repeated Alternating Stair Touch Test (RASTT, n = 37) where participants tapped, alternating right and left, with the ball of each foot a centered piece of tape on the top front edge of a small step as fast as safely possible for 20 s, and 3) the modified-RASTT (n = 37) which differed from the RASTT with the foot taps aimed straight forward on the step. Observations (# of steps for the RASST tests, and time (s) for SIP) were compared between two experimenters rating one trial from each participant using independent, two-tailed t-tests, Pearson (r) correlations, and intra-class correlations (ICC) for the two RASTT tests. Given its non-normal distribution, a Spearman Rank (rho) test and a two-tailed Mann-Whitney U test were used to compare the SIP test results between raters. RESULTS: There was no difference between raters for the RASTT (mean [SD]: 22.9 [6.8] vs. 22.8 [6.9] steps, p = 0.194), the modified-RASTT (mean [SD]: 23.3 [7.3] vs. 23.2 [7.2] steps, p = 0.168), or the SIP (median [IQR]: 10.7 [8.6] vs. 10.8 [8.6] s, U = 912, p = 0.920). The two experimenters’ ratings were highly correlated for both the RASTT (r = 0.998, ICC = 0.998), the modified-RASTT (r = 0.994, ICC = 0.993), and the SIP (Spearman’s rho = 0.997). CONCLUSION: Our results indicate that – while other analyses from this dataset indicate concerns regarding the repeatability of these stepping performance tests over two testing sessions – they do exhibit high inter-rater reliability