Influence of age, gender and test conditions on the reproducibility of Dual-Task walking performance.

Background: The review of methodological problems (confounding factors) of gait analysis in intervention studies with seniors is underrepresented.

Aim: This study focusses on two common problems of gait analysis under single-task (ST) and dual-task (DT) conditions (visual verbal Stroop test): (1) reproducibility of walking variables and (2) the effects of gait velocity, gender and age on peak plantar pressure to identify confounding effects on relevant outcome parameters.

Methods: The participants (N = 86, 71.9 ± 4.6 years) were divided into a (1) reproducibility (n = 28) and an (2) outcome parameter group (n = 58). Gait kinematics (step length; cadence) and kinetics (peak plantar pressure under heel, midfoot and forefoot) were analyzed walking barefoot on a treadmill (100 Hz) at self-selected speed for the reproducibility and at two different speeds (v = 3.5; 4.5 km/h) for outcome parameters. ICC analysis combined with the repeatability coefficient and SEM calculation, an ANOVA with repeated measurements and determination of effect sizes (η ) as well as a partial correlation analyses with body mass were done.

Results: The reproducibility of the walking variables under ST and DT conditions was excellent with ICC values of .67 to .99. The SEM and CR results as presented in Table 2 support these findings for some of the parameters.

Discussion: Plantar pressure values were influenced by gait velocity but less by age and gender. For DT walking the differences between preferred and fixed gait speed have to be controlled to assign the DT effects.

Conclusion: Effects of intervention studies should be carefully interpreted regarding the absolute reproducibility.