Background: The capacity to dual-task is critically important over the lifespan, enabling an individual to respond to demands in their environment, both safely and efficiently.

Research question: Does recent evidence suggest that relative to younger adults, older adults are most disadvantaged when performing locomotor-cognitive dual-tasks under conditions that are more representative of the real-world?

Method: A literature search of major electronic databases was conducted to find relevant peer-reviewed papers published since 2011. Thirty-nine studies that compared proportional dual-task costs (pDTC) between older and younger adults on a locomotor-cognitive dual-task were included. Study quality was assessed using the Appraisal tool for Cross-Sectional Studies.

Results: pDTC were calculated for a total of 504 motor and 53 cognitive outcomes. Weighted means showed that older adults experienced larger pDTCs than younger adults for motor (mean difference = -6.97) and cognitive (mean difference = -8.15) outcomes. Velocity variability measures produced the largest group difference on motor pDTC (mean difference = -32.83), as did cognitive tasks that targeted arithmetic (mean difference = -18.57) and texting skills (mean difference = -17.43). Cognitive tasks that were 'most representative' resulted in the largest age differences on motor pDTC (mean difference = -16.89). SIGNIFICANCE: This meta-analysis showed that dual-tasking challenged the ability of older adults to maintain consistency in the sequential timing of their gait. As well, older adults demonstrated greater pDTCs on motor outcomes, especially when the cognitive tasks were more representative of day-to-day activities. Taken together, this suggests that clinical assessments should focus on measures of variability rather than absolute measures of temporal and spatial gait. It is recommended that future research use more representative paradigms that are sensitive to dual-task interference and predictive of real-world behaviour.