Dual-Task Gait and Incident Dementia

Abstract

To the Editor We read with interest the study by Montero-Odasso et al 1 that shows that low dual-task gait speed is a sensitive marker of imminent dementia onset among patients with mild cognitive impairment (MCI). The study has important implications for dementia risk stratification among patients with MCI, which may help to select patients with MCI for putative neuroprotective clinical trials that aim to delay the clinical onset of dementia. However, 3 important gaps in knowledge remain.First, a critical question remains as to whether dual-task gait also predicts dementia in adults who do not yet have overt cognitive impairment. At that stage, pathological processes are typically less advanced among individuals who are prone to develop dementia, and putative neuroprotective interventions may still have substantial effects on the delay of dementia onset. Because its measurement is rapid and noninvasive, dual-task gait may therefore contribute to the selection of cognitively asymptomatic individuals for inclusion in neuroprotective intervention clinical trials.Second, while individuals with a high dual-task cost during the counting task had the highest hazard ratio for dementia in this study, it is noteworthy that individuals with a high "dual-task cost" during 2 relatively more complex tasks (naming animals or serial sevens) also had an increased risk of dementia. 1 While the latter 2 associations were not statistically robust in this relatively small sample, they question the extent to which the observed "dual-task cost" depends on the specific task that is administered. In fact, it is plausible that the task-specificity of the dual-task cost could vary by cognitive performance because complex dual tasks may be better at identifying high-risk individuals for dementia among cognitively asymptomatic populations than in patient populations with MCI. Therefore, a comprehensive assessment of the dual-task cost associated with several specific tasks among a cognitively heterogeneous population is warranted.Third, while this study focuses on the dual-task cost of gait speed, to our knowledge, the predictive use of the dual-task cost of other quantitative gait aspects for dementia remains largely unknown. In this context, a particularly promising gait aspect is variability (eg, high variability in stride length and time). Even under single-task conditions, variability is associated with cognitive performance and incident dementia independently of gait speed. 2,3 Under dual-task conditions, gait variability changes disproportionally more than gait speed does, and that difference is more distinct among patients with MCI than among individuals without cognitive impairment. 4 These findings highlight the need for a thorough investigation of the predictive use of quantitative gait aspects, including variability, for dementia.In the coming years, prospective cohort studies will accrue sufficient follow-up for dementia to robustly assess the usefulness of dual-task gait in predicting dementia among individuals without cogn...