Gait – the way we walk – starts slowing approximately 12 years before dementia diagnosis, suggesting that it reflects underlying changes in the brain and could indicate cognitive impairment.

The most common types of dementia, Alzheimer’s disease (AD) and Lewy Body disease (LBD), have different cognitive impairment profiles. However, these are difficult to detect in early stages and LBD is often misdiagnosed. As discrete gait impairments are associated with specific cognitive domains, we investigated whether these subtypes had unique gait signatures and could be an accurate diagnosis tool.

What did we do?

First, we examined gait in people with different dementia subtypes and older adults without dementia. We found that people with LBD have a unique gait signature compared to AD[1], and could tell apart people with any type of dementia from healthy older adults.
Next, we assessed inexpensive wearable technology in people with AD, using a small device placed on the lower back. This detected similar signatures of gait impairment between LBD and AD[2]. Using this approach in clinical practice as part of the diagnostic toolkit could reduce outpatient visits, healthcare costs and improve inclusivity through remote assessment of older adults – a need also highlighted by Covid-19.
This work was carried out as part of an Alzheimer’s Society funded PhD project conducted at the NIHR funded Clinical Ageing Research Unit, supported by a multi-disciplinary team. Our finding that gait impairment is a supportive indicator for diagnosis of prodromal LBD is cited in recent diagnostic criteria[3].

Next steps

We will develop a clinical decision-making tool to support differential diagnosis, involving three interlocking steps:
1. Enhancing models to differentiate dementia subtypes, by applying machine-learning to our wearable technology signals.
2. Collaborating with clinicians, people with dementia and carers to develop a clinical decision-making tool and data visualisation system.
3. External validation of the tool in patients with prodromal LBD, AD and healthy controls. Finally, assessing the feasibility of implementing the tool into practice and as a stratification tool for trials.

Linked Publications

1. Mc Ardle, et al., Do Alzheimer’s and Lewy body disease have discrete pathological signatures of gait? Alzheimer’s & Dementia, 2019
2. Mc Ardle, et al., Differentiating dementia disease subtypes with gait analysis: feasibility of wearable sensors? Gait & Posture, 2020.
3. McKeith, et al., Research criteria for the diagnosis of prodromal dementia with Lewy bodies. 2020.