Motor-cognitive effects of a computerized game-based training method in people with dementia: a randomized controlled...

To examine the effects of a computerized, game-based training on motor-cognitive performances, the transfer of training effects on untrained tasks, and the sustainability of training gains in people with dementia.

Ninety-nine individuals with a mean age of 82.9 (5.8) and dementia participated in a 10-week randomized controlled trial with three-month follow-up. The intervention group (IG) received a motor-cognitive training on (Physiomat®) including concurrent dual-tasks of balance control with cognitive demands (Physiomat®-Trail Making Tasks (PTMTs)). The control group (CG) performed non-specific, low-intensity exercises. Duration and accuracy at different complexity levels of trained and untrained PTMTs and the number of successfully performed tasks (PTMT score) were assessed.

Physiomat® training significantly improved the duration and accuracy at almost all complexity levels of trained (P ≤ 0.001-0.047, η_p² = 0.065-0.589) and untrained PTMTs (P < 0.001-0.005, η_p² = 0.073-0.459). Significant effects were also found for the PTMT score of trained (P < 0.001, η_p² = 0.211) and untrained PTMTs (P < 0.001, η_p² = 0.184). Training gains were partly sustained at follow-up.

Physiomat® is feasible and has the potential to sustainably improve motor-cognitive performances in people with dementia.