Meta‐analysis of type and complexity of a secondary task during walking on the prediction of elderly falls

Abstract

Dual-tasking probes divided attention and causes performance changes that are associated with an increased risk for falls in the elderly. There is no systematic review investigating the effect of task type and complexity on the prediction of elderly falls. This article synthesizes research evidence regarding this issue on the contents of dual-tasking walking. Relevant studies were systematically identified from electronic databases of Medline, PubMed, CINAHL, Cochrane CENTRAL and PsycINFO, and the reference lists of identified articles. The selection criteria were defined a priori. Two independent reviewers classified task types based on properties for cognitive demand, assessed the methodological quality with a customized checklist, and calculated the odds ratio of fall prediction. There was one study of reaction time, one of discrimination and decision-making, 10 of mental tracking, three of verbal fluency and five of manual tasks. The methodological heterogeneity was manifested in the selection criteria, faller classification, tasks and measures, resulting in substantial heterogeneity (I(2) 87-92%). Meta-analyses resulted in a significant pooled odds ratio 1.33 (95% CI 1.18-1.50). The mental tracking task was the only type that yielded a significant odds ratio 3.30 (95% CI 2.00-5.44). Running meta-analyses separately for simple and difficult mental tracking task showed similar odds ratios. The mental tracking task yielded significant dual-task-related changes for fall prediction. Most studies successively used an appropriate level of task complexity specific to the specified population of interest. More research is required for definite conclusions regarding the effect of task type and complexity.