Assessing walking speed in persons with a lower-limb amputation: Test-retest reliability of the 10-m walk test.

Background: Preferred walking speed (PWS), maximal walking speed (MWS), and walking speed reserve (WSR)-the difference between MWS and PWS-can be easily obtained from the 10-m walk test (10MWT) to assess walking ability and function. However, their test-retest reliability has not been determined in persons with unilateral lower-limb amputation (LLA).

Objectives: To determine the reliability of the PWS, MWS, and WSR obtained from the 10MWT in persons with LLA.

Study design: Test-retest with a 48- to 120-h interval between test days.

Methods: Test-retest relative and absolute reliability was assessed by the intraclass correlation coefficient (ICC) and the %change in mean (%CM) and coefficient of variation (CV), respectively.

Results: Preferred walking speed showed excellent reliability between day 1 (familiarization day) and day 2 (ICC = 0.97; %CM = -1.8; CV = 7.7) and a significant speed increase from day 2 to day 3 (ICC = 0.96; %CM = 8.5, p < 0.003; CV = 7.4). Maximal walking speed showed excellent reliability between day 1 and day 2 (ICC = 0.94; %CM = 2.9; CV = 8.3) and between day 2 and day 3 (ICC = 0.94; %CM = 1.8; CV = 8.6). Overall, WSR was poorly reliable between day 1 and day 2 (ICC = 0.65; %CM = 39.2; CV = 39.2) and between day 2 and day 3 (ICC = 0.74; %CM = 30.0; CV = 30.0).

Conclusions: Preferred walking speed obtained from a single 10MWT and MWS obtained after a 10MWT familiarization day are reliable walking speed measures for persons with LLA with similar characteristics to those of our sample. Nonetheless, it is critical that both PWS and MWS are obtained after a 10MWT familiarization to obtain a reliable WSR.