Gait, the way that we walk, requires complex cognitive functions. Gait may be a useful early marker for dementia diagnosis, as gait impairments precede and reflect cognitive decline. Early diagnosis of dementia enables individuals and their families to make informed decisions about their care plans, and allows researchers to understand preclinical and prodromal disease stages, providing novel targets for drug therapies. As such, a range of biomarkers are being developed to improve early and accurate diagnosis, including gait analysis. This editorial will outline how gait analysis can support the clinical diagnosis of dementia, including evidence of unique signatures of gait which can aid the identification of cognitive impairment and discrete dementia disease subtypes, the potential use of wearable technology to assess gait in the clinic and the real world, and key recommendations for the future implementation of gait into the diagnostic toolkit for dementia.
Prevalence of dementia is rapidly rising, with an estimated 131 million people worldwide expected to be diagnosed with the condition by 2050 . Dementia is characterised by multiple progressive cognitive impairments which impair an individual’s functional abilities . There is currently no cure, but both pharmacological and non-pharmacological therapies may be effective at treating key symptoms.
Developing methods for earlier and more accurate diagnosis is a research priority . This allows researchers to better understand the disease in early and predementia stages, providing novel targets for drug therapies. It also allows people with dementia and their families to make timely decisions surrounding their disease management, power of attorney and future care plans. There are a range of novel biomarkers being developed to improve the way we diagnose dementia, such as cerebrospinal fluid and blood markers, and neuroimaging correlates . However, many of these options are costly and invasive, and may not be feasible to employ wide-scale.
Inexpensive and non-invasive screening tools which detect subtle disease-related markers are required to support clinical decision-making. The advent of wearable technology, ambient sensors and smartphone applications have led to the development of digital biomarkers, which collect data relating to everyday functions, such as gait and sleep patterns, physical and social activity, and cognitive functions [5,6]. This may prove a fruitful endeavour to support dementia diagnosis.
How can gait analysis support clinical diagnosis of dementia?
Gait refers to the manner in which a person walks. Gait analysis is gaining significant interest as a digital biomarker [7-10]. Although gait appears automatic, it actually relies on significant cognitive processes . Gait impairments may be considered a red flag for cognitive decline and neurodegeneration, with a significant slowing of gait speed preceding dementia diagnosis by up to fourteen years .
Gait can be assessed within clinical settings through clinical examination, motor rating scales and self-report subjective scales such as the Tinetti Performance-Orientated Mobility Assessment or the Berg Balance Scale. These are commonly applied and useful for detecting change over time, but are limited by subjective variations and difficulty in identifying subtle gait impairments, which may not be observable to the human eye . Simply measuring gait speed can provide a more objective and sensitive assessment, but is not specific enough to characterise different diseases or subtypes of the same disease. However, there is more to gait analysis than assessing gait speed; there is a rich range of gait characteristics that can be examined , and which provide useful information regarding cognitive impairment, disease progression and efficacy of therapeutics in a variety of neurological conditions . Traditional gait assessment methods, such as instrumented walkways, commonly record spatiotemporal gait characteristics, such as those relating to pace (e.g. gait speed, step length), variability (i.e. changes in spatiotemporal characteristics of gait, such as how much step length changes across steps), rhythm (i.e. temporal characteristics of gait, such as step time), asymmetry (i.e. differences between left and right steps) and postural control (i.e. characteristics that contribute to keeping individuals upright while walking). These characteristics provide a more holistic picture of gait, allowing us to identify unique signatures of gait in dementia. For example, a systematic review reported that people with dementia take shorter step lengths and demonstrate greater variability compared to older adults without dementia .
Gait analysis may also be an effective differential marker of dementia disease subtype . Recent findings reported that people with Lewy body disease demonstrated a more variable and asymmetric gait compared to Alzheimer’s disease, suggesting Lewy body disease has a unique signature of gait that may reflect their underlying discrete pathology [17,18]. There have also been different gait patterns reported between Alzheimer’s disease and “non-Alzheimer’s dementias”, suggesting gait may be useful as a differential marker for subtypes such as vascular or frontotemporal dementia [19,20]. These subtypes of dementia may have similarities in clinical presentation and underlying pathology and thus be difficult to distinguish initially, leading to inaccurate diagnosis and inappropriate disease management . Inexpensive tools to support differential diagnosis of dementia subtype are required both clinically and in research to ensure the accurate characterisation of patients’ or participants’ disease subtype to ensure they are treating or researching the correct pathological targets . Additionally, accurate characterisation of disease subtype in early stages can support research into the identification of putative disease-modifying therapies.
How can clinicians measure gait in people with dementia?
Traditional gait analysis
Much of the evidence surrounding gait’s potential as an early and differential marker of dementia and its subtypes are derived from traditional gait assessments, such as instrumented walkways . This proves a barrier to uptake of quantitative gait assessment in clinic, as traditional gait analysis techniques require dedicated space, specially-trained staff and significant time to set up . However, wearable technology may provide a solution to inexpensively and easily assess gait in the constrained space of the clinic, with several algorithms validated to analyse gait in clinical settings .
Wearable technology for gait analysis
Wearable technology comprises of singular or multiple sensors (e.g. accelerometers – a device that senses motion and velocity, inertial measurement units – a device with multiple types of sensors which measures the position or orientation, force and angular rate of the body) which can be attached to the body or embedded within clothing or accessories, and which capture information and measurements relating to individual’s function, such as gait. Due to their inexpensive nature, they have garnered significant interest for clinical applications such as their utility as diagnostic tools for Parkinson’s disease . With the call for digital biomarkers to support dementia diagnosis , wearable-based gait analysis may be a feasible tool to aid the clinician’s toolbox.
Using wearable technology to assess gait in people with dementia has been shown to be feasible in both traditional laboratory settings and within the real world . People with dementia and prodromal stages of dementia have demonstrated significantly impaired gait patterns when assessed with wearable technology in the lab, such as slower gait, shorter steps, longer times to complete a step and greater gait variability [26-28]. One study has even demonstrated the efficacy of wearable-derived signatures of gait for differentiating dementia disease subtypes, with moderate-to-good accuracy for distinguishing dementia with Lewy bodies from Alzheimer’s disease, and good accuracy for differentiating Parkinson’s disease dementia from Alzheimer’s disease . This supports the proposal that wearable-based gait assessment may be a supportive tool for identifying cognitive impairment and differentiating dementia subtypes.
Analysing gait in the real world: potential for clinical use?
However, gait assessment in a lab or clinic only provides a snapshot of an individual’s gait performance, generally showcasing their best possible performance. Because of wearable technology’s ability to be worn continuously for prolonged time periods , we can now acquire a more realistic picture of an individual’s true gait performance, such as how they move and adapt to varying environments . There is significant interest regarding the use of wearable-based real-world gait analysis to monitor and detect neurological conditions [13,23,31,32]. Pilot studies have demonstrated that real-world gait impairments demonstrate good discriminatory accuracy to differentiate people with dementia from healthy older adults . However, lab-based gait assessment may be better at distinguishing between dementia subtypes, as differences in signatures of gait impairment are only found in very short ambulatory bouts in the real-world . Further work is required to explore the efficacy of real-world gait assessments for differential diagnosis of dementia. Additionally, there are a number of limitations to real-world gait which need to be addressed before implementing such assessments into clinical practice. These include the lack of consensus on which metrics best describe gait in the real-world (e.g. spatiotemporal or frequency based metrics), and the need for further validation of algorithms for detecting gait in the real-world as most have been developed for lab-based controlled environments . In light of the COVID-19 pandemic, it is clear that developing digital tools, such as wearable technology and its outcomes, is vital for healthcare services, given the necessity of remote clinical practice and diagnostic assessments .
What steps need to be taken to use gait as a diagnostic marker for dementia?
Digital biomarkers and assessment tools are not yet validated for clinical practice, and there are a number of steps that must be taken before they will be implemented for widespread clinical use. In line with In line with Rochester’s  roadmap for the development and implementation of digital outcomes, such as those produced via wearable-based gait assessment, we propose the following suggestions for researchers, clinicians and regulatory bodies to consider.
Firstly, we need to move beyond small pilot studies to maximise our data and understanding of how best to assess gait in the real world. Collaborative efforts are required to synthesise and standardise current gait assessment protocols, data processing and digital mobility outcomes . Research should incorporate the voices of clinicians, scientists, patients and regulatory bodies to ensure that the information produced by wearable-based gait analysis is relevant and improves clinical decision-making beyond what can be achieved through careful neurological examinations .
Secondly, wearable-based algorithms for gait analysis have been designed for controlled indoor settings, and do not account for changes in the environment that are frequently experienced in the real world (e.g., moving around a cramped room compared to walking outside) . This makes interpretation of gait impairments difficult, as we cannot identify which gait impairments are due to disease or due to the environment an individual is walking within . Criterion validity should be demonstrated by comparing digital outcomes to gold-standard references (such as traditional gait assessment) in both the lab and real-world, and assessing the feasibility, acceptability and usability of wearable technology with both clinicians and people with dementia. Similarly, construct validity should be demonstrated through longitudinal studies that assess how digital outcomes reflect more traditional measures of disease, such as cognitive and functional assessments .
Third, efforts to validate wearable-based algorithms and digital outcomes for identifying dementia, differentiating dementia subtypes and monitoring disease progression should be submitted to relevant regulatory bodies (e.g. European Medicines Agency) for approval.
Finally, cost-effectiveness of using wearable technology as a diagnostic tool must be assessed and demonstrated. Of particular importance here, quick and easy-to-use pipelines for processing and interpreting digital outcomes must be developed with the intention to implement widely. As an ideal example, the busy clinician would upload the data to a platform which would quickly process the outcomes and provide information regarding how likely the patient is to have a specific dementia subtype based on their gait patterns.
To summarise, we can see that gait assessment may be a useful way to identify cognitive impairment and distinguish dementia subtypes. With advances in wearable technology, it is easier to assess gait in the clinic and the real-world, which may be informative and supportive to the diagnostic process, thus serving as a digital biomarker. However, further work is required to validate wearable-based gait assessment as a diagnostic tool for dementia, including recruitment to larger collaborative studies, improved algorithm development, standardised protocols and digital mobility outcomes, and inclusion of patients, healthcare professionals and regulatory bodies in the research pathway. With the validation of wearable technology, we open up a myriad of possibilities, including more accurate diagnosis and monitoring of disease progression, increased confidence in stratification for clinical trials, and the ability to continuously and remotely assess responses to drug therapies or behavioural interventions, improving the potential for person-focused tailored care.
1. Prince M, et al. Dementia UK: -overview. 2014.
2. McKhann GM, et al. The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s & dementia. 2011;7(3):263-269. https://doi.org/10.1016/j.jalz.2011.03.005
3. Kenigsberg PA, et al. Dementia beyond 2025: Knowledge and uncertainties. Dementia (London), 2016;15(1):6-21. https://doi.org/10.1177/1471301215574785
4. Korolev IO, Symonds LL, Bozoki AC. A.s.D.N. Initiative, Predicting progression from mild cognitive impairment to Alzheimer’s dementia using clinical, MRI, and plasma biomarkers via probabilistic pattern classification. PloS one, 2016;11(2):p.e0138866. https://doi.org/10.1371/journal.pone.0138866
5. Kourtis LC, Regele OB, Wright JM, Jones GB. Digital biomarkers for Alzheimer’s disease: the mobile/wearable devices opportunity. NPJ digital medicine. 2019;2(1):1-9. https://doi.org/10.1038/s41746-019-0084-2
6. Piau A, Wild K, Mattek N, Kaye J. Current state of digital biomarker technologies for real-life, home-based monitoring of cognitive function for mild cognitive impairment to mild Alzheimer disease and implications for clinical care: systematic review. Journal of medical Internet research. 2019;21(8): p.e12785. https://doi.org/10.2196/12785
7. Mc Ardle R, Morris R, Wilson J, Galna B, Thomas AJ, Rochester L. What can quantitative gait analysis tell us about dementia and its subtypes? A structured review. Journal of Alzheimer’s Disease. 2017;60(4):p.1295-1312. https://doi.org/10.3233/JAD-170541
8. Beauchet O, Allali G, Launay C, Herrmann F, Annweiler C. Gait variability at fast-pace walking speed: a biomarker of mild cognitive impairment? The journal of nutrition, health & aging. 2013;17(3):235-239. https://doi.org/10.1007/s12603-012-0394-4
9. Rosano C, Snitz BE. Predicting dementia from decline in gait speed: are we there yet? Journal of the American Geriatrics Society. 2018;66(9):1659. https://doi.org/10.1111/jgs.15368
10. Beauchet O, Allali G, Montero-Odasso M, Sejdić E, Fantino B, Annweiler C. Motor phenotype of decline in cognitive performance among community-dwellers without dementia: population-based study and meta-analysis. PloS one. 2014;9(6):p.e99318. https://doi.org/10.1371/journal.pone.0099318
11. Morris R, Lord S, Bunce J, Burn D, Rochester L. Gait and cognition: Mapping the global and discrete relationships in ageing and neurodegenerative disease. Neurosci Biobehav Rev. 2016;64:326-45. https://doi.org/10.1016/j.neubiorev.2016.02.012
12. Beauchet O, et al. Poor gait performance and prediction of dementia: results from a meta-analysis. Journal of the American Medical Directors Association. 2016;17(6):482-490. https://doi.org/10.1016/j.jamda.2015.12.092
13. Buckley C, Alcock L, McArdle R, Rehman RZU, Del Din S, Mazza C, Yarnall AJ, Rochester L. The Role of Movement Analysis in Diagnosing and Monitoring Neurodegenerative Conditions: Insights from Gait and Postural Control. Brain Sci. 2019;9(2):34. https://doi.org/10.3390/brainsci9020034
14. Lord S, Galna B, Verghese J, Coleman S, Burn D, Rochester L. Independent domains of gait in older adults and associated motor and nonmotor attributes: validation of a factor analysis approach. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2013;68(7):820-827. https://doi.org/10.1093/gerona/gls255
15. Valkanova V, Ebmeier KP. What can gait tell us about dementia? Review of epidemiological and neuropsychological evidence. Gait & Posture. 2017:53:215-223. https://doi.org/10.1016/j.gaitpost.2017.01.024
16. van Iersel, MB, Hoefsloot W, Munneke M, Bloem BR, Olde Rikkert MGM. Systematic review of quantitativeclinical gait analysis in patients with dementia. Zeitschrift für Gerontologie und Geriatrie. 2004;37(1):27-32. https://doi.org/10.1007/s00391-004-0176-7
17. Fritz NE, et al. Motor performance differentiates individuals with Lewy body dementia, Parkinson’s and Alzheimer’s disease. Gait Posture. 2016;50:1-7. https://doi.org/10.1016/j.gaitpost.2016.08.009
18. Mc Ardle R, Galna B, Donaghy P, Thomas A, Rochester L. Do Alzheimer’s and Lewy body disease have discrete pathological signatures of gait? Alzheimers Dement. 2019;15(10):1367-1377. https://doi.org/10.1016/j.jalz.2019.06.4953
19. Allali G, et al. Gait phenotype from mild cognitive impairment to moderate dementia: results from the GOOD initiative. European Journal of Neurology. 2016;23(3):527-541. https://doi.org/10.1111/ene.12882
20. De Cock AM, Fransen E, Perkisas S, Verhoeven V, Beauchet O, Vandewoude M, Remmen R. Comprehensive quantitative spatiotemporal gait analysis identifies gait characteristics for early dementia subtyping in community dwelling older adults. Frontiers in Neurology. 2019;10:313. https://doi.org/10.3389/fneur.2019.00313
21. Kane JPM, et al. Clinical prevalence of Lewy body dementia. Alzheimers Res Ther. 2018;10(1):19.
22. Godfrey A, Brodie M, van Schooten KS, Nouredanesh M, Stuart S, Robinson L. Inertial wearables as pragmatic tools in dementia. Maturitas. 2019;127:12-17. https://doi.org/10.1016/j.maturitas.2019.05.010
23. Del Din S, Godfrey A, Mazza C, Lord S, Rochester L. Free-living monitoring of Parkinson’s disease: Lessons from the field. Mov Disord. 2016;31(9):1293-313. https://doi.org/10.1002/mds.26718
24. Gold M, et al. Digital technologies as biomarkers, clinical outcomes assessment, and recruitment tools in Alzheimer’s disease clinical trials. Alzheimers Dement (N Y). 2018;4: 234-242. https://doi.org/10.1016/j.trci.2018.04.003
25. Mc Ardle R, et al. Gait in Mild Alzheimer’s Disease: Feasibility of Multi-Center Measurement in the Clinic and Home with Body-Worn Sensors: A Pilot Study. J Alzheimers Dis. 2018;63(1):331-341. https://doi.org/10.3233/JAD-171116
26. Xie H, Wang Y, Tao S, Huang S, Zhang C, Lv Z, Wearable Sensor-Based Daily Life Walking Assessment of Gait for Distinguishing Individuals With Amnestic Mild Cognitive Impairment. Frontiers in Aging Neuroscience. 2019;1(285). https://doi.org/10.3389/fnagi.2019.00285
27. Mulas I, Putzu V, Asoni G, Viale D, Mameli I, Pau M. Clinical assessment of gait and functional mobility in Italian healthy and cognitively impaired older persons using wearable inertial sensors. Aging clinical and experimental research. 2020: 1-12.
28. Hsu YL, Chung PC, W, Wang WH, Pai MC, Wang CY, Lin CW, Wu LH, Wang JS. Gait and balance analysis for patients with Alzheimer’s disease using an inertial-sensor-based wearable instrument. IEEE journal of biomedical and health informatics. 2014;18(6):1822-1830. https://doi.org/10.1109/JBHI.2014.2325413
29. Mc Ardle R, Del Din S, Galna B, Thomas A, Rochester L. Differentiating dementia disease subtypes with gait analysis: feasibility of wearable sensors? Gait Posture. 2020;76:372-376. https://doi.org/10.1016/j.gaitpost.2019.12.028
30. Orendurff MS, Schoen JA, Bernatz GC, Segal AD, Klute GK. How humans walk: bout duration, steps per bout, and rest duration. Journal of Rehabilitation Research & Development. 2008;45(7). https://doi.org/10.1682/JRRD.2007.11.0197
31. Moore SA, Hickey A, Lord S, Del Din S, Godfrey A, Rochester L. Comprehensive measurement of stroke gait characteristics with a single accelerometer in the laboratory and community: a feasibility, validity and reliability study. Journal of neuroengineering and rehabilitation. 2017;14(1):130. https://doi.org/10.1186/s12984-017-0341-z
32. Moon Y, McGinnis RS, Seagers K, Motl RW, Sheth N, Wright JA Jr, Ghaffari R, Sosnoff JJ. Monitoring gait in multiple sclerosis with novel wearable motion sensors. PLoS One, 2017;12(2):p.e0171346. https://doi.org/10.1371/journal.pone.0171346
33. Gietzelt M, Wolf KH, Kohlmann M, Marschollek M, Haux R. Measurement of Accelerometry-based Gait Parameters in People with and without Dementia in the Field. Methods Inf Med. 2013;52(04):319-325. https://doi.org/10.3414/ME12-02-0009
34. Mc Ardle R, Del Din S, Donaghy P, Galna B, Thomas AJ, Rochester L. The Impact of Environment on Gait Assessment: Considerations from Real-World Gait Analysis in Dementia Subtypes. Sensors. 2021;21(3):813. https://doi.org/10.3390/s21030813
35. Rochester L, et al. A Roadmap to Inform Development, Validation and Approval of Digital Mobility Outcomes: The Mobilise-D Approach. Digital Biomarkers. 2020;4(1):13-27. https://doi.org/10.1159/000512513
36. Teipel S, Konig A, Hoey J, Kaye J, Kruger F, Robillard JM, Kirste T, Babiloni C. Use of nonintrusive sensor-based information and communication technology for real-world evidence for clinical trials in dementia. Alzheimers Dement. 2018;14(9):1216-1231.
37. Twardzik, E, Duchowny K, Gallagher A, Alexander N, Strasburg D, Colabianchi N, Clarke P. What features of the built environment matter most for mobility? Using wearable sensors to capture real-time outdoor environment demand on gait performance. Gait Posture. 2019;68:437-442. https://doi.org/10.1016/j.gaitpost.2018.12.028
38. Khandelwal S, Wickstrom N. Evaluation of the performance of accelerometer-based gait event detection algorithms in different real-world scenarios using the MAREA gait database. Gait Posture. 2017;51:84-90. https://doi.org/10.1016/j.gaitpost.2016.09.023
39. Del Din SKC, Yarnall AJ, Rochester L, Hausdorff JM. Body-worn Sensors for Remote Monitoring of Parkinson’s disease Motor Symptoms: Vision, State of the Art, and Challenges Ahead. Journal of Parkinson’s Disease. 2021. https://doi.org/10.3233/JPD-2024712.