The effects of verbal instructions on gait in people with Parkinson's disease: a systematic review of randomized and non-randomized trials.

Objective: To collate and appraise empirical evidence relating to the effects of verbal instructions (verbal commands given by another person) on stride length, gait velocity and stride variability in people with Parkinson's disease.

Data Sources: Cinahl, Cochrane, Embase, Medline, PEDro, PsycINFO and Web of Science.

Review Methods: Independent reviewers extracted data from eligible studies and assessed methodological quality. The level of evidence was determined by best evidence synthesis based upon the experimental design, methodological quality and statistical findings of individual studies.

Results: One randomized controlled study and 12 non-controlled studies fulfilled the selection criteria and involved 149 participants. Five types of verbal instructions were examined which included 'take big steps', 'walk fast', 'swing arms when walking', 'count rhythm when walking' and 'walk fast with big steps'. Best evidence synthesis found indicative evidence in support of the use of the instruction to take big steps in walking training for stride length improvement in people with mild to moderate Parkinson's disease who are without cognitive impairment. There was insufficient evidence in support of effects on gait velocity and stride variability. There was also insufficient evidence in support of effects of other instructions on any of the gait variables.

Conclusion: The empirical evidence in support of the benefits from verbal instructions is weak. The evidence is limited to short-term stride length improvement from the use of the instruction to take big steps in walking training.