Table 3.
Summary of reviewed articles.
| Article |
Participants
N mean age* % female Initial cognitive state |
Time of the follow-up |
DT paradigm (motor task + cognitive task) |
Cognitive assessment at baseline and follow-up | DT parameters |
|---|---|---|---|---|---|
| Åhman et al. (2020a) | 17271.0 years 45.3% MCI or SCI people from a memory clinic |
2.5 years | TUG + naming animals TUG + reciting months backward |
MMSE, 7 Min Screen neurocognitive test (specifically Clock Drawing and Verbal Fluency tests). | • Simple TUG time (s) • Naming animals DT time (s) • Naming animals DT time cost (%) • Numbers of words recited in naming animals DT (n) • Words recited per 10 s during the naming animals DT (n) • Counting backward DT time (s) • Counting backward DT time cost (%) • Number of words recited in the reciting months backward DT (n) • Words recited per 10 s during the reciting months backward DT (n) |
| Beauchet et al. (2017) | 56 68.9 years 46.4% Community-dwelling older people without dementia |
5 years (4.8 ± 0.7 years) | UGS + naming animals UGS + counting backward since 50 |
MMSE. | • Initial stride time characteristics: • Mean value and CoV in usual gait • Mean value, CoV; delta mean value, and delta CoV in counting backward DT • Mean value, CoV; delta mean value, and delta CoV in verbal fluency DT |
| Ceïde et al. (2018) | 1,156 78.3 years 60.7% Community-dwelling older adults without dementia |
1.9 mean years | UPG + reciting alternate letters of the alphabet (instructed paying attention equally in both tasks) | Short form of Wechsler Adult Intelligence Scale-Revised (WAIS-R), Mill Hill Vocabulary scale, FCSRT, BIMC. | • DT parameters: • Speed (cm/s) • Cadence (step/min) • Step length (cm) • Swing percentage (%) • Stance percentage (%) • Swing time SD • Step time SD • DT gait domains: • Rhythm • Variability • Pace |
| De Cock et al. (2019) | 433 80.0 years 45.7% Older people from a memory clinic with healthy cognition, MCI, and incident dementia |
5 years | UPG + naming animals UPG + counting down from 50 in steps of two |
MMSE, ACE, NPI-Q, CDR. | • From usual pace gait, fast-paced gait, slow-paced gait, naming animals DT gait, counting backward in steps of two DT gait were obtained the following variables: • In pace gait domain: • Gait speed (cm/s) • Normalized gait speed*** (m/s) • Cadence (steps/min) • Steps/meters (steps/m) • Step length (m) • Normalized steps per meter*** (steps/m) • Postural control gait domain: • Step width (cm) • Step width variability (%) • Variability gait domain: • Swing time variability (%) • Cycle time variability (%) • Additionally, from both DT paradigms were obtained: • DT gait speed cost (%) • DT cadence cost (%) • DT step width cost (%) • DT width step variability cost (%) • DT cycle time variability cost (%) • DT swing time variability cost (%) • DT steps/meter cost (%) • Normalized DT steps/meter cost*** (%) • Counted numbers during 10 meters in counting backward DT (discrete number) • Counted animals during 10 meters in naming animals DT (discrete number) |
| Deshpande et al. (2009) | 660 74.6 (5.3) 54.2% Community-dwelling older people |
3 years | UPG + naming animals | MMSE | • Gait speed (m/s) of: • UPG • Fast-paced gait • DT gait |
| Donoghue et al. (2018) | 2,250 72.4 years** 52.0%** Community-dwelling older people without dementia |
5.9 years | UPG + reciting alternate letters of the alphabet | MMSE, Verbal Fluency, Immediate recall, Delayed recall, MoCA, Color trails 1 time, Color trials 2 times, Color trials time difference, Cognitive response time, Movement time, Total response time, SART mean response time (ms), SART SD (ms), SART coefficient of variation (%), SART errors of commission (n), SART errors of omission (n) | • TUG time (s) • UPG speed (cm/s) • DT speed (cm/s) |
| Gillain et al. (2016) | 13 73.1 years** 46.2%** MCI older people from a memory clinic |
4 years |
UPG + counting backward from 50 |
MMSE, CDR. | • Variables obtained from UPG and DT: • Gait speed (m/s) • Regularity (absolute value) • Symmetry (absolute value) |
| Montero-Odasso et al. (2017) | 112 76.0 years 49.1% Community-dwelling older people with MCI |
2 years (12–76 months) |
UPG + naming animals UPG + counting backward from 100 UPG + subtracting several sevens from 100 |
MMSE, CDR, TMT-A, TMT-B, Rey auditory verbal Learning test, Boston naming test, Digit Span Forward, Digit Span backward, Letter-Number Sequencing test. | Single task gait speed (cm/s, continuous variable) • Counting backward from 100 DT gait speed (cm/s, continuous variable) • Subtracting several sevens from 100 DT gait speed (cm/s, continuous variable) • Naming animals DT gait speed (cm/s, continuous variable) • Counting backward from 100 DT gait speed cost (%, continuous variable) • Subtracting several sevens from 100 DT gait speed cost (%, continuous variable) • Naming animals DT gait speed cost (%, continuous variable) • Lower gait speed in the single-task (<0.8 m/s, dichotomic variable) • High cost in Counting backward from 100 DT gait speed (>20%, dichotomic variable) • High cost in subtracting several sevens from 100 DT gait speed (>20%, dichotomic variable) • High cost in Naming animals DT gait speed (>20%, dichotomic variable) |
| Muurling et al. (2020) | 142 67.0 years 47% Older people from a memory clinic grouped in healthy cognitively, MCI, and mild dementia. |
1.2 years** (1–2 years) | WT8 + counting backward from 100 | MMSE, CDR. | • Variables of UPG and DT: • Mean stance time (s) • Mean stride time (s) • Mean swing time (s) • Cadence (steps/min) • Stance time variability (s) • Stride time variability (s) • Swing time variability (s) • Mean step length (m) • Speed (m/s) • Step length variability (m) |
| Nielsen et al. (2018) | 86 72.0 years 20.7% Older people from a memory clinic grouped in cognitively healthy, MCI, and mild dementia |
2.5 years** (12–36 months) | TUG + counting backward from 100 | MMSE, ACE, CDR. | • TUG time (s) • TUG DT time (s) • TUG DT cost (%) • DT performance (normal, moderate deviation, and severe deviation) |
| Osuka et al. (2020a) | 626 76.0 years 61.8% Community-dwelling older people without cognitive impairment |
2 years | S-TMT | MMSE, TMT A. | • S-TMT time divided in tertials: highest, middle and lowest (s) • TMT-A time (s) • UPG speed (m/s) |
| Rosso et al. (2019) | 223 78.7 years 52.5% Community-dwelling older people without cognitive impairment |
9 years | UPG + visuospatial clock task | 3MS. | • UPG speed (0.1 m/s) • Fast-paced gait speed (0.1 m/s) • Narrow path gait speed (0.1 m/s) • DT gait speed (0.1 m/s) • Fast-paced gait speed cost (%) • Narrow path gait speed cost (%) • DT gait speed cost (%) |
This table summarizes the characteristics of the 12 reviewed prospective studies. Each row is representative of a study. The first column indicates the participant's characteristics (number, sex, age, and cognitive status); the 2nd column indicates the follow-up time in years; the 3rd column summarizes the used DT tasks (motor + cognitive); 4th column summarizes the cognitive assessment performed in each study, and the last column summarizes the main dual-task gait variables that are used as a predictor of cognitive impairment in older adults. DT, dual-task; DT performance: Categories are 3: “Normal” (no notable changes in either gait velocity or performance of the cognitive task during gait); “Moderate deviation” (dual-tasking imply changes in either gait velocity or performance or performance of the cognitive task dual-task); “Severe deviation” (dual-tasking imply either detention in gait when engaging the cognitive task or incapability of performing, according to Nielsen et al., 2018). Motor tasks, TUG: Timed up and Go test; UPG, Usual pace gait; WT8, Walking test (consisting in standing up from a chair, walking 5 m and going around a cone, coming back and going around the chair, the cone again, and sitting down in the chair, forming an 8, at a comfortable, fast pace); S-TMT, Stepping Trail Making Test (consisting of a test where the participant is asked to walk inside a 1 × 1m square divided into 16 squares of the same size numbered from 1 to 16 in an established order, stepping the squares in consecutive order) (Osuka et al., 2020b). Cognitive tests: Visuospatial clock task: It is a test where the participant is asked to indicate if the clock's needles are in the same or the opposite half of the clock, respecting a line between 12 and 6. MMSE, MiniMental State Scale; CDR, Clinical Dementia Rating scale; MoCA, Montreal Cognitive Assessment; SART, Sustained Attention Response Task; TMT, Trail Making Test; FCSRT, Free and Cued Selective Reminding Test; BIMC, Information-Memory-Concentration test; ACE, Addenbrooke's cognitive examination; NPI-Q, Neuropsychiatric Inventory Questionnaire; 3MS, Modified Mini-Mental State; rs, Spearman's rank correlation coefficient; CI, confidence interval; OR, Odd Ratio; HR, Hazard Ratio; MCI, Mild Cognitive impairment; SCI, Subjective cognitive impairment; AD, Alzheimer's disease.
Ages and female percent were rounded to one decimal
This information was calculated indirectly.
Adjusted by leg length. DT cost was calculated from the formula: [(usual pace gait parameter – gait parameter)/usual pace gait parameter] × 100. CoV: Coefficient of variation was calculated = [(standard deviation/mean) × 100]. Delta mean value, and delta CoV were calculated from the formula: [dual-task – single task/(dual-task + single task)/2] × 100. Delta MMSE was calculated from the formula: [baseline MMSE – MMSE at 5 years of follow-up/(baseline MMSE + MMSE at 5 years of follow-up)/2] × 100.