Gait disorders and slowed walking speeds may be useful indicators of future cognitive decline


The way people walk is an indicator of how much their brains, as well as their bodies, are aging.

Scientists reporting in a special supplement to the Journal of Alzheimer’s Disease (JAD) say that gait disorders, particularly slowing gait, should be considered a marker of future cognitive decline.

They propose testing motor performance as well as cognitive performance in older adults with mild cognitive impairments.

“There is an emerging focus on the importance of assessing motor performance as well as cognitive performance to predict cognitive function loss,” explained guest editor Manuel Montero-Odasso, MD, PhD, Departments of Medicine (Geriatric Medicine), and Epidemiology and Biostatistics, Schulich School of Medicine & Dentistry, University of Western Ontario, and Gait and Brain Lab, Parkwood Institute, Lawson Health Research Institute, London, ON, Canada.

“In the past two decades, large epidemiological studies have shown that gait disorders, particularly slowing gait, may be present at early stages of dementia or may even predict who will be at risk of progressing to dementia.

Subtle impairments in gait are more prevalent in older adults with cognitive impairments and dementia and are also associated with an increased risk of falls.”

This supplement presents a comprehensive assessment of the current state of knowledge on the role of gait disturbances and quantitative gait analyses to be used as a motor biomarker to define subtypes of cognitive profiles and predict cognitive decline and dementia.

It also evaluates screening tools that are important for developing strategies to prevent or delay onset of clinically meaningful change.

The issue showcases studies on:

  • Epidemiology of gait disturbances and cognitive impairment
  • Association between gait speed and further cognitive decline
  • Structural brain volume covariance and its association with gait speed in patients with mild cognitive decline
  • Association of cerebral amyloid-beta deposition and impaired gait speed and lower extremity function
  • Use of the dual-task gait (DTG) paradigm of walking while performing a concurrent cognitively demanding task as a brain stress test to detect populations at risk
  • Value of DTG to detect individuals at risk of cognitive decline in a cognitively healthy population of older adults
  • Effect of learning to use a mobility aid on gait and cognitive demands in AD – the use of assistive devices can help improve mobility and reduce the risk of falls but can also present a cognitive challenge
  • Feasibility of measuring gait in an outpatient cognitive neurology setting

Gait testing may help to detect the subgroup of at-risk patients who may benefit the most from invasive diagnostic procedures or early interventions.

“We believe simple gait testing should be part of routine clinical assessment for older adults with cognitive impairments.

Implementing this in clinics may be a challenge, but we hope the evidence presented in this issue will lead to progress in this area,” noted guest editor George Perry, PhD, Editor-in-Chief of JAD, Professor of Biology, Semmes Distinguished University Chair in Neurobiology, The University of Texas at San Antonio, TX, USA.

This shows an older man walking

There’s something in the way you walk. The image is credited to School of Medicine & Dentistry, University of Western Ontario, et al.

“Finding early dementia detection methods is vital,” added Dr. Montero-Odasso. “It is conceivable that in the future we will be able to make the diagnosis of AD and other dementias before people even have significant memory loss.

In older adults with moderate cognitive impairment, slowing down their usual walking by more than 20 percent when they add a cognitive task is indicative of a seven-fold increased risk to develop AD in a five-year timeframe.

We believe that gait, as a complex brain-motor task, provides a golden window of opportunity to detect individuals at higher risk of dementia who can benefit the most from more invasive testing or early interventions.”

Cognitive impairment and dementia carry a heavy healthcare burden. Around fifty million people worldwide have dementia. Every year there are nearly ten million new cases.

AD is the most common form, accounting for around 60-70 percent of cases. Dementia is characterized by a progressive loss of cognitive function affecting memory, thinking, orientation, comprehension, calculation, learning capacity, language, and judgement. Impairments in gait are more common in dementia than in normal aging and may be related to the severity of cognitive decline.

In 2015, an estimated 46.8 million people worldwide had dementia.1 Dementia is a contributor to disability and life years lost in older individuals.2 There is no cure for dementia, so identifying potential risk factors may reveal opportunities for prevention. One area of interest is whether physical function is related to dementia onset, because declines in physical and cognitive functioning are indicators of aging, and gait disorders increase with age3 and are associated with incident dementia.4

Walking speed is easier to assess than other gait parameters. Slow walking speed is associated with negative outcomes in older individuals.567 Individuals with cognitive impairment and dementia walk more slowly than individuals without these conditions 8 Furthermore, meta‐analytical evidence indicates that slow walking speed is a predictor of dementia.49 Change in walking speed in relation to dementia risk has been less well researched than current walking speed. But in 3,663 French adults, those with a steeper decline in walking speed were at greater risk of dementia than those with a slower decline.10 Similar findings have been reported in Swedish and Japanese samples.1112

Dementia develops slowly and can be preceded by years of decline in cognitive functioning.13 Cognition and physical function influence one another in a complex manner.14 There is some evidence that the association between cognition and walking speed is bidirectional,15 although most studies find that slow walking is a predictor of decline in cognition but not vice versa.1617

The relationship between changes in cognition and walking speed has been assessed in several studies, but the results are equivocal.181920 Findings from a sample of 762 participants in the MacArthur Studies of Successful Aging showed that cognition and walking speed declined in tandem over a 7‐year period.18 A limited association was also detected in a Tasmanian sample, in which a decline in executive function (but no other domains) was associated with a decrease in walking speed.19 An analysis from the Women’s Health Initiative Memory Study failed to detect any association.20

Overall, it appears that slow walking speed9 is associated with greater dementia risk, with more‐limited evidence that faster decline in walking speed is also relevant.101112 The evidence is mixed as to whether changes in cognition are associated with changes in walking speed,181920 and these associations have not been examined in relation to dementia risk. To address these questions, we evaluated whether walking speed and change in walking speed were predictive of dementia in a sample of 3,932 English adults. We also evaluated whether changes in cognition and walking speed have an interactive effect on dementia risk.

Participant Characteristics

We compared the characteristics of those who developed dementia (n= 289) with the characteristics of those who did not (n=3,643). Those who developed dementia were significantly older on average, were less wealthy, had poorer cognition and mobility and significantly slower walking speed, and were more likely to have had a stroke or depressive symptoms than those who did not (Table ​(Table11).

Table 1

Participant Characteristics (2002–04) According to Dementia Status (2006–15)

CharacteristicDementia, n = 289No Dementia, n = 3,643P‐Value
Age, n (%)<.001
60–6963 (21.8)2,016 (55.3)
70–79135 (46.7)1,248 (34.3)
≥8091 (31.5)379 (10.4)
Male, n (%)110 (38.2)1,621 (44.5).06
Wealth, decile, mean5.34 ± 0.175.85 ± 0.05.004
Education, n (%).70
<Junior high school125 (43.2)1,512 (41.5)
High school99 (34.3)1,293 (35.5)
University65 (22.5)838 (23)
Walking speed, m/s, mean ± SD0.78 ± 0.020.87 ± 0.01<.001
Cognition, mean ± SD (range −2.97 to 4.02, Z score)–0.32 ± 0.04–0.06 ± 0.01<.001
Mobility impairment, mean ± SD (range 0–10)a2.44 ± 0.141.95 ± 0.04.001
Activity of daily living difficulty, mean ± SD (range 0–6)b0.41 ± 0.040.32 ± 0.01.05
Coronary heart disease, n (%)50 (17.4)532 (14.6).16
Stroke, n (%)22 (7.5)149 (4.1).02
Diabetes, n (%)25 (8.8)270 (7.4).26
Hypertension, n (%)132 (45.6)1,537 (42.2).16
Cancer, n (%)16 (5.6)189 (5.2).67
Center for Epidemiologic Studies Depression Scale score, mean ± SD (range 0–8)1.83 ± 0.111.38 ± 0.03<.001

All analyses are age‐adjusted.aDifficulty with ≥1 common arm and leg functions (range 0–10).bWhether participants had difficulties with 6 daily activities (range 0–6).

SD = standard deviation.

Baseline Walking Speed as a Predictor of Dementia

Walking speed at Wave 1 was a predictor of development of dementia, with those with faster walker speeds being less likely to develop dementia (HR=0.36, 95% CI=0.22–0.60) during follow‐up (Table ​(Table2).2). This association was robust to adjustment for covariates. Cognitive function at baseline was also an independent predictor of development of dementia (HR=0.42, 95% CI=0.33–0.52), with those with better cognitive function being less likely to develop dementia.

Table 2

Hazard Ratio of Walking Speed (2002–03) and Dementia Incidence (2006–15)

FactorAdjusted HR (95% Confidence Interval)P‐Value
Walking speed, m/s0.36 (0.22–0.60)<.001
Cognitiona0.42 (0.33–0.52)<.001
Sex1.09 (0.84–1.42).48
Age (reference 60–69)
70–792.91 (2.14–3.94)<.001
≥806.68 (4.70–9.49)<.001
Wealth, decile0.96 (0.92–1.01).09
Education (reference <junior high school)
High school1.01 (0.76–1.33).960
University1.24 (0.87–1.77).242
Mobility impairmentb0.98 (0.92–1.05).596
Activity of daily living difficulty1.03 (0.87–1.21).762
Coronary heart disease1.12 (0.83–1.52).446
Stroke1.26 (0.82–1.93).295
Diabetes mellitus1.11 (0.74–1.65).614
Hypertension1.06 (0.84–1.34).626
Cancer1.03 (0.63–1.70).894
Depression1.06 (0.99–1.13).087

a1–standard deviation increment in cognition associated with adjusted hazard ratio (HR).bDifficulty with ≥1 common arm and leg functions (range 0–10).

Changes in Walking Speed and Cognition as Predictors of Dementia

Walking speed decreased on average from 0.86 to 0.85 m/s between Waves 1 and 2. Change in walking speed was a significant predictor of dementia (HR=1.23, 95% CI=1.03–1.47), with those who had a greater decrease in walking speed from Wave 1 to Wave 2 having a greater risk of developing dementia independent of covariates and walking speed in Wave 1 (Table ​(Table3).3). Change in cognition was also a predictor, with participants with a greater significant decline in cognitive function between Waves 1 and 2 being at greater risk of developing dementia (HR=1.78, 95% CI=1.53–2.06), but the interaction between walking speed and cognitive function was not a significant predictor of dementia (HR=1.01, 95% CI=0.88–1.17).

Table 3

Hazard Ratio of Change in Walking Speed and Cognition (from 2002–03 to 2004–05) and Dementia Incidence (2006–15)

FactorAdjusted HR (95% Confidence Interval)P‐Value
Change in walking speed, m/s1.23 (1.03–1.47).02
Change in cognition, z scorea1.78 (1.53–2.06)< .001
Walking by cognition interaction1.01 (0.88–1.17).88
Baseline walking speed, m/sb0.28 (0.14–0.57)< .001
Baseline cognitionb0.27 (0.21–0.35)< .001
Female1.19 (0.89–1.56).23
Age (reference 60–69)
70–793.03 (2.17–4.22)< .001
≥806.59 (4.49–9.68)< .001
Wealth (decile)0.97 (0.92–1.02).28
Education (reference <junior high school)
High school0.96 (0.71–1.31).80
University1.44 (0.97–2.13).07
Mobility impairmentc0.99 (0.92–1.06).68
Activity of daily living difficulty0.99 (0.82–1.20).90
Coronary heart disease1.21 (0.87–1.67).26
Stroke1.04 (0.63–1.70).88
Diabetes mellitus1.18 (0.77–1.81).45
Hypertension0.96 (0.75–1.24).77
Cancer1.08 (0.62–1.90).79
Depression1.03 (0.96–1.10).48

a1–standard deviation increment in cognition associated with adjusted hazard ratio (HR).b1‐m/s increment in walking speed associated with adjusted HR.cDifficulty with ≥1 common arm and leg functions (range 0–10).

Sensitivity Analyses

Three sensitivity analyses were conducted. First, we tested whether walking speed at Wave 2 (2004–05) predicted development of dementia. Participants with faster walking speeds were less likely to develop dementia (HR=0.25) between Waves 3 and 7 (2006–2015) (Supplementary Table S1).

Second, we only considered new events from Wave 4 to 7, omitting any that occurred within 2 years of Wave 2. This reduced the number of cases from 289 to 225 (Supplementary Table S2), but walking speed at Wave 1 remained a predictor of dementia (HR=0.33). Similarly, those with poorer cognition at Wave 1 (HR=0.29) and those who had a greater decline in cognitive function between Waves 1 and 2 (HR=1.69) remained more likely to develop dementia. The effect size for changes in walking speed was similar to that in the full analysis, but the association was no longer significant because of the smaller number of cases.

For the final sensitivity analysis, we excluded participants who were diagnosed using the IQCODE and limited the analysis to those with physician diagnoses. This reduced the number of cases from 289 to 240 (Supplementary Table S3). Walking speed remained a predictor of dementia (HR=0.36). The findings for cognition also remained for baseline cognition (HR=0.26) and change in cognitive function (HR=1.83), but similar to the other sensitivity analysis, change in walking speed no longer predicted development of dementia.Go to:


In this sample of 3,932 older adults, those with a slower walking speed were at greater risk of developing dementia. Individuals who experienced a greater decline in walking speed were also at greater risk. Participants with poorer cognition at baseline and those who experienced greater decline in cognition were also more likely to be diagnosed with dementia, although change in walking speed and change in cognition did not have an interactive effect on development of dementia.

Our finding that slower walking speed is a predictor of subsequent dementia is in line with previous research.49 In a meta‐analysis of 10 prospective studies in this area, 9 were conducted in the United States and one in Sweden.9 Our study adds to this literature by demonstrating an association between walking speed and dementia in an English sample. To the best of our knowledge, this is also the largest sample used to address this question. The association remained significant in our sensitivity analyses, demonstrating the robustness of this finding.

The association between change in walking speed and dementia risk has not been investigated as much. In the current study, those with greater decline in walking speed over 2 measurement periods were more likely to develop dementia. This finding is in agreement with research in Swedish, French, and Japanese cohorts.101112 The relationship between change in walking speed and dementia remained after controlling for numerous covariates, including baseline walking speed. This suggests that, regardless of initial walking speed, a marked decrease over a relatively short period (2 years) may be an indicator of greater dementia risk.

We were interested in how cognition and walking speed might influence one another in affecting dementia risk. First, we found that individuals who had poorer baseline cognition were more likely to develop dementia. Second, those with greater decline in cognitive function between Waves 1 and 2 were at higher risk of dementia over follow‐up. Finally, we investigated whether changes in cognition and walking speed interacted to affect risk of development of dementia, but we found no evidence to support this.

Initial cognitive function and decline in cognitive ability are associated with greater risk of subsequent dementia,13 but understanding of the interplay between changes in walking speed and cognition is limited. Previous research attempting to ascertain whether these functions decline in parallel has been inconsistent, with one study suggesting that these functions decline in tandem,18 another finding an association but only for one component of cognition (executive function),19 and another finding no association at all.20

The present study extends this research by investigating whether changes in cognition and in walking speed interact to predict future dementia. Our findings suggest that declining walking speed and declining cognitive function are independent predictors of dementia but that these factors do not work synergistically. Cross‐sectional evidence suggests that gait speed maps onto the stage of cognitive impairment; with the fastest speeds reported in individuals with mild cognitive impairment and the slowest speeds reported in those with advanced dementia.28 It may be that this interplay between cognition and walking speed emerges only at the stage of clinically significant impairment rather than when individuals are still cognitively healthy.

This observational study provides information on the chronological, but not the causal, relationship between walking speed and dementia. Furthermore, reverse causality might be a factor. There is weak evidence from some1516 but not all studies17 that cognitive function may predict walking speed and that walking speed might influence later cognitive status. In our sensitivity analysis excluding cases that occurred in the first 6 years after the walking speed assessment, the association between walking speed and subsequent dementia remained. This addresses the notion that undetected dementia cases affected walking speed and adds weight to the temporal sequence that gait slowing precedes dementia onset.

With regard to the mechanisms linking walking speed with future dementia, several possibilities could help explain our findings. It is thought that walking and cognition rely on similar brain regions, predominately in the prefrontal cortex.2930 Gait is a complex process in which the locomotor systems receive input from the basal ganglia, motor cortex, and cerebellum.30 Although this process is largely automatic, walking relies on sensory feedback and high‐order cognitive control.2930

Neurodegeneration is a possible underlying mechanism linking declines in physical and cognitive function, with changes in subcortical white matter31 and cortical gray matter volumes31 associated with slower gait speeds. Vascular risk factors may also contribute to the link between gait speed decline and dementia through a similar pathway, because micro damage to the vessels of the prefrontal cortex and lesions (e.g., from stroke) is associated with white matter changes,32 but our analyses were robust to adjustment for vascular risk factors and history of cardiometabolic disease.

Another potential mechanism is low‐grade systemic inflammation. High concentrations of inflammatory markers are predictive of new‐onset dementia33 and have also been implicated in mobility impairment.34 Neuroinflammation is thought to lead to impaired neuroplasticity in the brain areas controlling motor and cognitive function.30 Furthermore, walking speed relies on muscle strength,35 and muscle loss has been linked with deleterious inflammatory processes.36

These findings may have implications for efforts to delay dementia onset. The effect of exercise interventions on cognitive function and dementia is unclear.3738 Nonetheless, a meta‐analysis of 42 studies examining the effects of 3 exercise interventions on walking speed39 suggested that exercise can lead to increases in walking speed of up to 9.3%. Exercise may also improve cognitive function. A 2017 meta‐analysis of 36 randomized control trials found that various types of exercise had a beneficial effect on cognition, regardless of baseline cognitive status.40

We analyzed data from a large representative sample of English adults using an objective test of walking speed and were able to confirm associations between walking speed and dementia risk after adjusting for multiple confounders, such as mobility impairment and cardiometabolic disease. Dementia was primarily identified according to self‐reported physician diagnoses, which resulted in fewer cases than would be expected based on population estimates.41 When this was supplemented by including diagnoses based on the IQCODE, the incidence of dementia was in line with U.K. estimates.41 Although it is possible that we missed cases, given the consistency of our findings with earlier studies,49101112 it is unlikely that misclassification bias accounted for our results. Furthermore, self‐reported diagnoses of other conditions are noted to correspond closely with physician diagnoses, even in individuals with cognitive impairment.42

The average decrease in walking speed between Waves 1 and 2 was small and may not be clinically significant. Although the findings of our study suggest that individuals with larger declines in speed are more likely to develop dementia. Some aspects of cognition are thought to be more strongly associated with physical functioning than others.1115 However, this study did not investigate associations between walking speed and different aspects of cognition, meaning that potential interactions before dementia onset may have been missed.

Overall, this study suggests that individuals with slower walking speeds and those who have a greater decline in walking speed over time are at greater risk of dementia. Changes in walking speed and in cognition did not interact to affect development of dementia, indicating that they operate through independent pathways. Further research is required to understand the causal mechanisms underlying these associations and to determine whether increases in walking speed reduce dementia risk.

IOS Press
Media Contacts:
Diana Murray – IOS Press
Image Source:
The image is credited to School of Medicine & Dentistry, University of Western Ontario, et al.

Original Research: The special edition of Journal of Alzheimer’s Disease is available online.


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.