DUAL-TASK INTERFERENCE DURING WALKING IN INDIVIDUALS POST-STROKE: A SYSTEMATIC REVIEW AND META-ANALYSIS

File
Tsang CSL1, Wang S2, Pang MYC1
1The Hong Kong Polytechnic University, Rehabilitation Sciences, Hung Hom, Hong Kong, 2Sun Yat-Sen University, Guangzhou, China

Background: Dual Task Interference (DTI), a decrement in task performance, may occur when a secondary task is done concurrently with walking (i.e., dual-task walking). However, dual-task walking is critical in effective community ambulation. In the past decade, an increasing number of studies have examined the effect of DTI during walking in persons post-stroke.The type and difficulty level of the mobility and secondary component tasks comprising the dual-task walking condition may have important influences on the degree and pattern of DTI. This study consolidates the available evidence on the phenomenon of gait-related DTI in individuals with stroke, and the effect of component task type and difficulty on DTI.

Purpose: To examine the degree and pattern of DTI during dual-task walking in individuals post-stroke, and how these vary with component task type and difficulty.

Methods: Systematic literature search of CINAHL, Cochrane Library, EMBASE, MEDLINE (1946+),PsycINFO (1806+) and PubMed were done. Observational or experimental studies written in English that reporteddual-task walking performance of individuals post-stroke were selected. The patterns and degree of DTI were examined. Meta-analyses were performed as appropriate to quantify the overall effect of DTI in people post-stroke. Sensitivity analysis was conducted to examine the DTI effect in different subgroups of people with stroke.

Results: Fifty of the 10,072 articles generated from the database search met the inclusion criteria for the review, and 35 for meta-analysis. Mental tracking was the most commonly adopted secondary task type in the dual-task testing paradigm. Dual-task gait speed was the most common outcome measure. Mutual interference (deterioration of performance in both component tasks) was the most common DTI pattern. Meta-analysis showed the greatest DTI on gait speed when a secondary mental tracking task (MD= -0.14m/s, 95%CI=-0.18, -0.10, p 0.01,I2= 45%)or motor task (MD= -0.14m/s, 95%CI= -0.18, -0.11, p 0.01, I2= 32%) was imposed. Within the mental tracking category, the greatest DTI on gait speed (MD= -1.34m/s, 95%CI=-2.01, -0.66, p 0.01,I2= 85%), cadence (MD= -1.53steps/min, 95%CI=-2.22, -0.84, p 0.01,I2= 58%)and stride length (MD= -1.17m, 95%CI=-2.33, -0.00, p=0.05,I2= 86%) was found during walking with serial-1-subtraction. Sensitivity analysesofDTI on gait speed in people with chronic stroke (i.e., six months post-strokeor more) showed similar results as the primary analyses.

Conclusion(s): Available evidence suggested more pronounced DTI on gait speed, cadence and stride time in people post-stroke when a mental tracking task was imposed, relative to other secondary component tasks. Within the mental tracking task category, serial-1-subtraction induced the strongest DTI.

Implications: The findings revealed that the DTI during walking is a substantial problem among people with stroke, and that the greatest deficit is related to the addition of a mental tracking task while walking. As ability to engage in dual-task walking is integral to community reintegration, there is a need to address dual-task walking in stroke rehabilitation. This systematic review has provided insight to the DTI phenomenon in people with stroke and should be useful in the development of dual-task exercise training programs to improve dual-task mobility function post-stroke.

Keywords: Dual-task walking, Stroke, Systematic review

Funding acknowledgements: The study was supported by a research grant provided by the Research Grants Council (grant no: 151594/16M).

Topic: Neurology: stroke; Disability & rehabilitation

Ethics approval required: No
Institution: The Hong Kong Polytechnic University
Ethics committee: Departmental Research Committee
Reason not required: This is a systematic review.


All authors, affiliations and abstracts have been published as submitted.

Back to the listing