|| Checking for direct PDF access through Ovid
Dual-task (DT) performance, the ability to divide one's attention between motor and secondary tasks, is required in daily life. Adults with cognitive impairment (CI) experience more difficulty with DTs than healthy older adults, but it is unclear how the degree of CI relates to DT performance, particularly with tasks of varying levels of difficulty.The purposes of this cross-sectional study were to (1) explore the relationship between cognitive level and DT performance and (2) determine how the difficulty of the combined tasks impacts this relationship.Twenty-three older adults with Mini-Mental State Examination (MMSE) scores ranging from 7 to 30 performed 2 single tasks (ST): the Timed Up and Go (TUG) and a 6-m walk for which self-selected walking speed (SSWS) was calculated. Each ST was repeated under 2 DT conditions: counting forward by 1's (TUG1 and SSWS1) and counting backward by 3's (TUG3 and SSWS3). Dual-task cost (DTC) was calculated for each DT as follows: [(difference between DT and ST motor performance)/ST motor performance] × 100. Spearman rank correlation coefficients were used to determine the relationship between DTC and the MMSE. The Friedman 2-way ANOVA on ranks was used to compare the magnitude of DTC among the 4 DTs.Significant correlations between the MMSE and DTC were found for SSWS3, TUG1, and TUG3 (r = 0.43-0.57). SSWS1 had a weaker and nonsignificant correlation between MMSE and DTC (r = 0.36). The TUG3 was the most difficult DT, while the SSWS1 was the easiest DT. All participants, regardless of MMSE score, were able to engage in all DTs.A linear relationship exists between cognition and DTC in older adults with varying cognitive levels. The strength of this relationship is greater for more challenging tasks. We also suggest that patients with CI may be able to engage in more challenging tasks than might be assumed. The impact of task difficulty has implications in the design of future studies of DT training for individuals both with and without CI.