Podcast – Gait and dementia

Voice Over:

Welcome to the NIHR Dementia Researcher Podcast, brought to you by dementiaresearcher.nihr.ac.uk., in association with Alzheimer’s Research UK and Alzheimer’s Society, supporting early career dementia researchers across the world.

Dr Anna Volkmer:

Hello. I’m Dr. Anna Volkmer. I’m a speech and language therapist and researcher based at University College London, and I’m very pleased to be guest hosting this podcast today for the NIHR Dementia Researcher website.

Dr Anna Volkmer:

Now, gait is a person’s pattern of walking, and walking involves balance and coordination of muscles so that the body is propelled forward in a rhythm. There are numerous possibles that may cause an abnormal gait. In my own research, I study language and communication as a means to help understand what might be happening in someone’s brain. In a similar way, researchers are looking at people’s gait and disorders that might affect it.

Dr Anna Volkmer:

Now today, I am joined by three researchers who are at the forefront of gait research, analysing gait and undertaking studies to see what it can tell us. I’m delighted to introduce Dr. Riona McArdle, Dr. Keir Young, and Dr. Silvia Del-din. Hello, everyone, and thank you for joining us today.

Dr Riona McArdle:

Hiya.

Dr Silvia Del-din:

Hello.

Dr Keir Yong:

Righto.

Dr Anna Volkmer:

Now, can I start by asking you to all introduce yourselves properly? So let’s start with Riona maybe.

Dr Riona McArdle:

Hi. So I’m Riona McArdle. I’m a research associate at Newcastle University, and I’m a psychologist by background. But now my research really focuses on the application of gait analysis and remote monitoring techniques to improve the way that we diagnose and care for people living with different types of dementia.

Dr Anna Volkmer:

Thank you, Riona. How about Silvia? Tell us a little bit about yourself.

Dr Silvia Del-din:

Hello, everyone. I am Dr. Silvia Del-din. I’m a bio engineer by background, and I’m a senior research associate at Newcastle University in the same group as Riona works in. My main research really focuses on the use of variable technologies, so for example, accelerometer sensors, response sensor, et cetera, for quantifying gait, walking and digital mobility outcomes. I work with both data collected in the laboratory environment but also, which is quite exciting, with real world data, so data collected in [inaudible 00:02:44] condition, at-home environments. I work with ageing population and also in neurodegenerative diseases like Parkinson’s disease, for example, and dementia.

Dr Anna Volkmer:

Natural walking, naturalistic walking. And last but not least, Keir.

Dr Keir Yong:

Thanks, Anna. So my name’s Keir Yong. I am a senior researcher fellow and Alzheimer’s Society fellow working at Dementia Research Centre, Queen’s Square Institute of Neurology. And my main area of work involves working with people living with various degrees of dementia-related visual impairments.

Dr Keir Yong:

So these refer to difficulties perceiving what or where things are, arising not from eye conditions, but rather diminished ability to interpret information caused by damage, particularly areas towards the back of the brain. An area of working also focuses on use of technology including sensors like inertial measurement units, or kind of motion capture, that technology to really evaluate care interventions to maximise independence of people living with various degrees of dementia-related visual impairment.

Dr Anna Volkmer:

Fascinating. Thank you very much. Now, just to start with, Silvia, can I come to you first and could you perhaps tell us a little bit more about what gait is? I don’t know if everybody uses the term “gait” a lot. So tell us what can go wrong, and how you can actually measure gait?

Dr Silvia Del-din:

Of course. So, gait is the way that we walk, that we move. And maybe not many of us know that gait is also referred as the sixth vital sign. So, yes. Like you would measure, for example, your blood pressure or your body temperature to understand how well you’re doing, your health, gait, so the way that we walk, can also serve as a tool for evaluating our health. Maybe we can relate also this concept to the fact that, as we age, we might think that we get older, we get slower, we walk with a lower velocity. And research, for example, has shown that gait, together with age, and specifically gait speed, so the velocity in which we are walking, can really predict life expectancy and then, so mortality. So if we look at research, gait is really being shown to be a biomarker. So a biomarker of ageing, as I was mentioning, but also of pathology. So a lot of different diseases like Parkinson’s disease, like for example, dementia, dementia sub-types, have shown specific gait pattern and impairment in their walking. And if we can think about it, gait is really our unique signature, our unique fingerprint. And so it can monitor also the way that we are progressing, whether it’s ageing, whether it’s a specific disease. I think that the relevant and important thing here is really thinking about gait not only as an automatic and mechanic activity, okay? People might just think, “Okay, we put foot in front of the other.” It’s not so simple. In fact, gait involves a lot of brain activity, if you want, and control of cognitive function as well. Everything starts from our brain, and after myself, I think Riona also will touch upon this, and Keir as well. Obviously, attention, decision-making is very important while we walk, because we’re not just walking. We might be walking and talking to someone. We might be walking and watching our smartphone. So, sometimes maybe people, they might have had some [inaudible 00:06:18] where they are distracted and maybe they are tripping on the curb, or maybe they’re even falling.

It’s not so funny for the person who experiences that. We might have seen videos of people falling while watching their mobile phone. This is because we’re multitasking, so when we’re distracted, so we take away the attention, the cognitive control, we might therefore have some gait problem of gait impairment as well. This is all very interesting, but if we do blood pressure, if we do measure our temperature, we use some instrument, okay? So, how can we actually measure gait and walking? I can start by saying that typically and traditionally, gait is measured in the lab.

So in the laboratory, a controlled environment, people come along and we use quite expensive and bulky, I would say, system. For example, maybe a motion capture system, so [inaudible 00:07:12] motion capture system, where you can use some reflective markers on your body, and with infrared cameras, it can really track the movement, and so quantify gait, if you want. I don’t know. Maybe people have seen the Lord of the Rings, where Gollum was … The actor was, he had a lot markers position on his face, and therefore with this camera, you can reconstruct, if you want, the movement. We do the same, but with gait samples. Other instruments, for example, are instrumented matrices. They are pressure sensor-based matrices, where you can walk on, and you can derive a lot of measure of gait. For example, how long your step is, how quickly you’re walking, for example, the rhythm of your step, et cetera. And it’s like really leaving your footprints on the sand, if you want, and then you can measure a lot of outcomes from gait. Obviously, all of these, so all the data that we can really get from the laboratory, can give us a massive amount of useful information that has been used in the past, as I say, for describe and really identify gait, and gait impairment in a variety of pathology and diseases. If we want to describe gait, within our lab, we have, or if you want to come up with a model of gait, that can be described in five different domains with our [inaudible 00:08:30] lab environment and very understandable.

The five domains are: Pace, so fast how we walk; rhythm, our rhythm of walking, for example, the step time, the time to take a step; variability, so what’s the difference between one step or the other? Asymmetry, so the difference between, for example, our right and left step; and also postural control, so for example, how wide our base of control is while walking, et cetera. This is all fascinating, but the real question is, if we measure gait, how we walk in the lab, is it a realistic picture of our behaviour, of our habitual gait, and the way that we walk? Does it possibly just capture maybe capacity, so what people can do, because we’re in front of them or asking them to walk normally in a certain way? And so there might be something called the [ozone 00:09:20] or white coat effect, where people might just not perform as they usually do, just because they’re in front of the doctor. As I was mentioned before, if you go to the doctor to measure blood pressure, it might be higher, not because you have a higher blood pressure, but because you’re under pressure and maybe you’re anxious of being in front of a doctor.

So, it’s very important for us to be able to measure gait and walking when people are in their everyday life, if you want, real-world environment. And we can do now, because with the new technology, we have a lot of wearable technology, wearable sensors that I was mentioning before, that can help us really measure the way that we walk in real life condition. We can still use them in the lab, and we can validate, and we can see whether they are really robust in quantifying, if you want, the gait characteristics that we know we can quantify in the lab. But then we can also use them outside of the lab, and get maybe more interesting information also about gait. So I think that’s an overview of gait.

Dr Anna Volkmer:

Yeah, that’s really helpful. You’ve gone over a lot of areas, and I really like the idea that gait is like a fingerprint, that kind of analogy. That’s really helpful. I think also, for any of our listeners who aren’t in the health and research field, that makes it really accessible.

Now, you’ve talked a little bit about some of the common problems that cause abnormal gait, but I actually wondered whether Riona, you could tell us a little bit about your research? Because I know that gait isn’t just one element of your research, but the main focus of your research.

Dr Riona McArdle:

Yeah, thank you. So, my research has really been born out of the Parkinson’s disease research that Silvia has quite a lot of expertise in. And there’s been a lot of work, as Silvia said, that has associated gait with cognition. So, when a person’s cognition begins to decline, we also see that their gait may actually predict that cognitive decline. For example, there was a meta-analysis published a few years ago, that suggested between 12 and six years before a person would research their diagnosis of dementia, their gait would actually begin to significantly slow down, and it might be a red flag for developing dementia.

And so there has been quite a lot of work that’s been unpacking this, showing that gait impairments are prevalent in dementia, even though we might not expect it to be a disorder that would have motor problems within it. However, there hasn’t been a lot of work done on what type of dementia a person might have, and if gait could be predictive of what type of dementia that person would have. And that’s really where my research has come into it.

So I’ve done this throughout my PhD, and I’ve continued to do it within my post-doc. So I’m really interested in two of the most common types of dementia: Alzheimer’s disease and Lewy body disease. And although these look very different on paper, when you actually see a person with Lewy body disease in clinic, sometimes it can be very hard to tell if they have Lewy body disease, or if actually they have Alzheimer’s disease. And so there’s a range of biomarkers that is trying to unpack this problem, and trying to find out, how can we definitely know if someone has Lewy body disease, or how can we be more confident that someone has Lewy body disease? The problem with a lot of these biomarkers, which are generally like imaging biomarkers, such as DAT scans for Lewy body disease, are that they’re quite expensive, and not everybody would receive one. So if someone really looks like they had Alzheimer’s disease, you probably wouldn’t send them on for a DAT scan to check if they’ve got Lewy body disease.

And so we really need to find a screening tool that can help us decide who we need to send on for further assessment. And that’s what I wanted to find out if gait analysis could be useful for. So as Silvia said, we’ve got lots of different methods to do this. I’ve used both the gold standard techniques, the instrumented walkway within the lab, and also the newer techniques using wearable technology, both in the lab and in the real world environments. And I’ve been trying to find out if Lewy body disease have a unique signature of gait impairment compared to Alzheimer’s disease. So, do they have a specific walking pattern that we could be confident in saying, “This is Lewy body disease”? And the reason that we are interested in the unique gait impairment pattern is because we know that discrete gait impairments are associated with specific cognitive domains. So for example, in the literature, we see that people who have attentional or executive dysfunction, they often walk slower, with greater variability of their gait.

They’re changing their steps up a lot more, in comparison to someone who doesn’t have attentional or executive problems. And so we thought, because there’s different cognitive profiles in Lewy body disease and Alzheimer’s, they would also have unique gait profiles that would reflect this. So, in the results from my work, we did find that Lewy body disease have a unique signature of gait impairment, using both gold standard techniques and wearable technology techniques within the laboratory environments. So people with Lewy body disease are a lot more variable when they’re walking. They change their step lengths a lot more and their step times a lot more. And interestingly, they’re also a lot more asymmetric when they’re walking, so their left and right footsteps look quite different from each other. And that was quite an interesting finding, because actually, they seem to be more asymmetric in their pathologies while in the early stages of disease. And this might give us a bit of an indication that actually, gait is reflecting what’s occurring in the brain, although we do need to do further work to validate this. And it also showed us that perhaps it might be a useful tool for a differential diagnosis of dementia sub types.

Not only was it useful for picking up differential diagnosis of dementia sub types, it was also useful at identifying just general cognitive impairment, in comparison to our age match controls who don’t have any cognitive impairment. So, people with Alzheimer’s and Lewy body disease all walked slower, with shorter footsteps. They had more variability in their walking, and they also took wider steps. So we can see that actually, it is a good feature of just picking out cognitive impairment in general, if that was what your question was.

Dr Anna Volkmer:

Yeah.

Dr Riona McArdle:

However, we did also look, as Silvia said, at gait in the real world, using wearable technology. So, we placed a small sensor on people’s lower backs and asked them to wear it for seven days, and they just went off into their environments and walked around as we expected.

And again, we could see that there was signatures of gait impairments that significantly differed the dementia sub types, but it was much harder to interpret in the real world environment. So for example, we found that we could differentiate Alzheimer’s and Lewy body disease when they were walking in very short walking bouts. So for example, if they were only moving for 10 seconds, or between 10 and 30 seconds. But once they got to longer, steady state walking in the real world environments, these differences were much less apparent. Now, that could be because I had a small sample size. We only had 125 people recruited into this study, so we had very small groups. But it could also just be, environmentally, we don’t really understand how gait is changing at the moment. And these studies are really unobtrusive, so we don’t know where they’re moving around, what they’re doing. We need to get a better understanding of that as we continue to go on. So, that’s kind of where I’m at, at the moment. I’ve got lots of plans for the future, but that’s sort of the state of the research that we’re at, at the moment, with me.

Dr Anna Volkmer:

And I love the idea that you’re using sensors on the backs to get that naturalistic sample. I think you probably need to put one on me, because I confess, I’m one of those people who falls over at least once or twice a year. And I generally attribute it to either a lack of concentration, or poor depth perception with my glasses on. Which kind of brings me to Keir, in a way, because Keir, you’ve been looking more at vision and perception, in terms of gait and dementia, haven’t you, in your work?

Dr Keir Yong:

Yes, that’s correct, Anna. So, as we were discussing earlier, I wouldn’t consider myself to primarily be a gait researcher. But rather, I’ve used gait measures to look at effect of clinical presentation. So I work with people who have a condition known as “posterior cortical atrophy,” that some people refer to as a visual variant of Alzheimer’s disease. But also to look at effects of environmental adaptions and conditions, again, on everyday walking, whether that’s in a controlled movement laboratory, but increasingly, we’re interested in moving into more everyday settings.

Now, where the role of the physical environment seems to be particularly important with people with more visual presentations, really arises from patient reports of things like people not being so confident when they’re walking over certain surfaces. So for example, with patterned surfaces, people overstepping perceptual variations in flooring. Patients mentioning things like glare or shadows being very disconcerting, and I’m talking about people who are experiencing some of these, in some cases, as their first symptoms of their underlying Alzheimer’s disease.

As well as really quite unsettling reports where people walk across reflective surfaces, and they mention things like it actually looks like there’s a sheer drop. But a lot of these reports also somewhat chime with anecdotes we hear from people at more advanced stages of more considered to be typical Alzheimer’s disease. So my first roles were actually working on care home settings. And again, when we did focus groups with some staff and you ask people, “Well, actually, what do some residents struggle with?” You get staff saying things like, “Well, actually, people seem to walk very hesitantly around certain parts of the home. People will avoid, say, this particular area where we have this garish carpet.” Also, people not really being able to judge how much clearance they have when they’re walking between furniture. In some cases, actually bumping into furniture. So, to return to some points raised by Silvia and Riona, so I’d like to start off by saying, again, in line with my background not really being in gait assessment, when I first started working in this area, I kind of oversimplified, I think, what you could actually derive from some of these movement sensors. So I thought, “How hard it can be, you know? You’re just going to attach some sensor on someone who’ll give you all the information you want. Acceleration, velocity, displacement, and position in GPS coordinates.” However, as Silvia’s mentioned, unfortunately it’s not quite that straightforward. However, I’ve worked with engineers on a technique called as “pedestrian dead reckoning.” Now, this is something that I understand actually people like emergency workers can use to keep track of their position, so let’s say you’re a firefighter walking throughout a burning building, and you’ve got a motion sensor.

It can give you not only acceleration and velocity, but you can also integrate what I deem “corrected velocity,” to estimate displacement, going back to a point of origin, which could be where you’re actually coming into that burning building. So, we’ve used this with people with, again, visual-led presentation of Alzheimer’s disease, and people with more memory-led presentation of Alzheimer’s disease, walking around an accessibility laboratory, snappily called the “pedestrian accessibility and movement environment laboratory.” But the advantage of this pedestrian dead reckoning technique is, technically, it’s an infrastructure-less technique. I.e., you don’t actually need a gait laboratory to use it. Now, an advantage of this technique is we can evaluate certain effects of environmental conditions. So for example, we can see about directness of paths taken to a destination under different lighting conditions. We can also look at gait variability, so we’ve worked with people from the London School of Hygiene and Tropical Medicine on this approach called “step time outlier detection,” where you can take into account an individual’s usual walking speed, or even their walking speed under a particular condition. So, let’s say we have high lighting variability, and you’ve got lots of shadow, versus low lighting variability, where you’re controlling the extent of shadows, and you can see where in that environment you might have disproportionately slow steps. And I think more widely, some of the interest we have, again, in some of these measures, is not only getting insight into gait disturbances, but looking at really specific aspects of, again, postural control, whether it’s on standing balance tasks, or again with things like transferring from standing to sitting.

Because for some patients I work with, not only those with this posterior cortical atrophy, but some people who have memory-led Alzheimer’s disease, but then quickly develop a lot of secondary visual spatial problems, so they’re missing where things are. They might also have a lot of difficulty with dressing. These symptoms can be really disabling, and interfere with things like reliable transfers.

Dr Anna Volkmer:

Absolutely, and that can create a massive risk. I know, having worked across hospitals, outpatients, care homes, transfers and falls is one of the first things that staff are anxious about and worried about. So that said, we started with Silvia. I’m actually going to come back round to Silvia, because what we didn’t talk about was specifically more about your research around Parkinson’s, for example. And I just wondered if you could share some of that with us?

Dr Silvia Del-din:

My research focuses on the implementation, if you want, creation of methods and analytics behind these technologies, for example, and wearable sensors. So, how do you get information from a signal? You need to create an algorithm. You need to create some sort of analytic that can get you that information. So that’s my real expertise and interest. Obviously, as mentioned also by Keir and Ri, we have a lot of wearable technologies, and a lot of them are medical commercial devices, so somehow black boxes.

So, we don’t really know how valid or good they are in quantifying. So the first thing I did in my research was, first, we needed to show how well with analytics that I was implemented, we could evaluate these gait parameters, these gait variables that we can derive from the gold standard, so for example, from the instrumented walkway that Riona was mentioning.

So the first thing I did was really a validation study, and that was in people with Parkinson’s disease, and in people that also have [inaudible 00:23:28]. And what we did was we compared the output that we could get from one sensor on the lower back, to the output that we can get from the gold standard. And we saw that there was quite a good validity, apart from … and we need to be honest, all the methods are not perfect. We could see some discrepancies or some variability in asymmetry metrics as well. But we saw that possibly the sensor is potentially more sensitive to this type of, if you want, slow movement or asymmetry in variability metrics, rather than the gold standard itself.

Anyway, once we did that, we were able to, if you want, and we were ready to utilise the sensor in real-world conditions. So then I moved into the real-world condition. And what we saw was that, in people with Parkinson’s disease, again, trying to identify some differences with this, with the control, we saw that in the lab, we could find some differences, for example, especially in the pace domain. So, people with Parkinson’s were walking with shorter steps and were slower. When we look at how they behave in the real-life condition, we saw that all the differences that we found in the lab were exaggerated. And we could find many more differences in real-life condition than compared to what we found in the lab. So, not only they were walking slower with shorter steps, but they were also walking in a more asymmetrical and variable way as well. So therefore, linking back to what I was saying at the beginning, maybe real-life conditions can really tell us much more of what we can see in the lab, or better can complement, if you want, what we can see in the lab or in the clinic.

Then after doing that, so this was really a snapshot, if you want, observation of how people with Parkinson’s can walk, what their gait impairment is. And lately, I’ve been moving towards the more prodromal phase of Parkinson’s disease. So we know that people, before actually being diagnosed with Parkinson’s, there is maybe 20 years where the disease might be starting, but the actual motor symptom, for example, or other symptoms that can allow a diagnosis, are not there yet. So people have the disease but they’re not diagnosed with that. So, I was able to collaborate with Professor Walter Maetzler from Kiel University, and also with Professor Michele Hu from Oxford University, in looking at two prodromal cohorts. And we found out that, even though in these prodromal stages, people that will possibly develop Parkinson’s, can also have already a gait impairment. So, something within their walking pattern and gait, that can tell us that possibly, they’re going to then develop, for example, Parkinson’s disease. And within the work with Professor Maetzler, we found out some specific gait characteristics, so gait speed, so the way that the velocity in which we walk, and again, step length, so the length of our step, can change and start changing for up to four years before actually being diagnosed.

And that was very, very interesting because it’s something that I presume is very fascinating. So if we are able to say, “Well, this is possibly a screening tool that we can use, and these are some red flags and ringing bell that can tell us that possibly someone is going to develop Parkinson’s, or is at risk of Parkinson’s, then we can maybe intervene someone, with intervention, with rehabilitation, with drugs, even when they will be available at the very beginning, and so prevent possibly also the disease to either appear or to obviously decline.” So, I think that now, that’s quite interesting, I would say. But again, it all goes back to the tools that we’re using. If I have to say, my main interest is really being able to create analytics that are quite strong [inaudible 00:27:20] so that researchers like Keir or Riona or Anna or whoever wants to use them, and just sort of show that they’re using and they have a valuable outcome. And out of this, we have been able to create, if you want, an online platform that allows people who want to collaborate with us, to actually, wherever they are in the world, use our protocols, upload the data, and get the information that we can provide.

So, in terms of walking behaviour, as Riona was mentioning, and in terms of gait characteristics, for example, as I was mentioning, gait speed, et cetera. And so, I don’t know. It’s just like, if people are interested, they can look up and get in touch.

Dr Anna Volkmer:

Absolutely. That’s a great idea. Look up and get in touch, and look at all the tools you’ve got on that resource. And as primarily a clinician, or a clinician before I became a researcher, in speech therapy, we’d often work with people with Parkinson’s. But often, we’d also work with people before they’d been diagnosed, and it’s fascinating. We’d have people referred for a voice issue, and they’d walk into the clinic, and there’d be no Parkinson’s.

And as soon as they walked in, the team would say, “That’s Parkinson’s.” Just from the way they walked into the room. And then as they opened their mouth and we’d hear the quality of their voice, that would often add to that and we’d say, “Yeah. Yup, probably Parkinson’s.” And then we’d refer them on. But it’s always fascinating that you can tell things from just watching someone. I confess, I am one of those people in the community, I’ll say to my husband, “I think that person might have …” because you can just tell by the way they’re walking around.

And I think gait’s becoming more and more of an area of interest, in fact, and the focus of maybe more than a dozen posters and presentations at the recent AAIC conference. And I recall a poster from [Margin Merling 00:29:20], I think, who found associations between gait disturbances with cognitive impairments, and neuro-degeneration, which is essentially part of what you’re looking at, Riona, isn’t it? So, my question now is, I wonder how this area is going to change and progress in the next 18 months. Do you have any opinions? Maybe coming to Riona first.

Dr Riona McArdle:

So, do you know what? I think this podcast has been really nice and complementary, because a lot of the things that I hope are going to be changing in the 18 months or over the next few years, have been touched on and complemented by both Silvia and Keir as well.

I think we are at the start with gait in dementia. There’s been a long path getting here already, but we still have quite a lot of work to do. But the work is getting ready to go. As I said, I’ve been really fortunate when I did my PhD, because I worked with Silvia and I also worked with Professor Lin [Rochester 00:30:15], who are at the forefront of gait in Parkinson’s disease. And that really provided me with a framework of what I needed to follow in order to look at gait as a diagnostic marker for dementia. So I think in the future, what we really need to do is, we need to have larger studies, and we need to do a large replication study that really includes a lot of biomarkers and follows people up post-mortem pathology, to allow us to have confidence in what our results are.

We need to think about these kind of biomarkers like imaging, and also blood biomarkers, but also not just that. Other things that we could look at as well, for example, Anna, you’re doing work with speech and language. And that might be a really complementary thing to come into, where gait sits within the best battery of things in the clinician’s toolkit.

I think the best way to do that, of course, is collaborative, multi-centre studies, so that we are not limited to one [inaudible 00:31:06] area. There is some studies up and coming within this field. The deep and frequent phenotyping study is looking at prediction of Alzheimer’s disease and people with the APOE-E4 gene, and that’s going to be including gait along with a whole kitchen sink of biomarkers as well. So we might be able to find some interesting results within that. We’ve already done a lot of the work in terms of finding a validated model with a comprehensive range of gait and assessing that within dementia, and different dementia sub-types, and examining the associations with cognition.

But really now, what we need to do is find out, what is the relationship with pathology? Can we be confident in saying, for example, step length variability is associated with pathology related to Lewy body disease? Is that something that we can actually say? And we need to find that out.

We’ve also found that wearables are feasible to use, and maybe effective to use in dementia and picking up what type of dementia a person has. But really, we need to be doing longitudinal work in prodromal cases. So, can we predict what type of dementia a person has? And do those gait characteristics progress and reflect the progression of cognitive impairment and increasing pathology? So I think there is quite a lot of work to be done, but we are getting there. And also, I remember when I published my paper in Alzheimer’s and Dementia on this, the kind of main paper for my PhD, a reviewer made a very good comment that I’m just looking at Alzheimer’s and Lewy body disease. There is many, many types of dementia.

And Keir is talking about PCA as his type of dementia. So phenotypes off these Alzheimer’s disease and so on, may have their own signatures of gait impairment that we’re not aware of yet. We need to look into that. But I do think that the work that has already been done has paved a foundation for people to continue to look at that.

Dr Anna Volkmer:

Yeah. I can see that with the devices, a collaboration between people wearing devices to measure gait, and also to somehow capture naturalistic communication and conversation? Because that’s one of the markers that we look at in speech therapy, across … Actually, my area of research is mainly the language-led dementias, which are primarily progressive aphasia’s, which also have much in common with some of the PCAs, the posterior cortical atrophies patients in terms of the symptoms and the language and communication difficulties. So, there’s so much work to be done, isn’t there? Keir, did you want to add anything else to that, in terms of what you feel will happen or needs to happen over the next 18 months?

Dr Keir Yong:

Thank you. So, just to add onto Riona’s point about phenotyping, so I think there’s quite an interesting area that really sits at the intersection between phenotype and, say, environmental characteristics, whether they’re physical or social. And of course, I’m slightly banging on about this because of work I’ve been involved in, but in a sense, when we look at gait response, we’re really looking at almost a gait response to visual information.

So terms that are a bit counter intuitive are things like an adapted gait response, or perhaps even just say, a visual motor function, but to do with how you really modulate walking in response to demands of the immediate physical environment. And I really hope that we’ll see development on this front, both in controlled settings, and ideally in community, everyday settings. Now, we’ve already had a great summary on the potential of gait measures and facilitating people getting to a timely and accurate diagnosis, and even as potentially in differential diagnosis, from Riona, in an area that I understand is particularly challenging in distinguishing Alzheimer’s disease from dementia with Lewy bodies.

The area I think we’re going to see, hopefully, some big developments in, is the use of measures like gait measures to build capacity in environments that are currently considered quite challenging for high quality care research. So for example, and I’m slightly shooting from the hip here, you might have some kind of intervention that’s intended to reduce wandering, whether that’s because it’s to manage agitation, but one of your outcomes, say your second outcome measure, is the length, total path, distance, cumulative distance travelled within certain time windows. I would like to see work in that area. Again, going back to when I was based in care home settings, a lot of our outcome measures for interventions like [inaudible 00:35:35] simulation therapy were questionnaire-based, and they have sub-optimal properties for a number of reasons, whereas there’s something that’s quite scalable about having wearable sensors. But for this to work, we need to have actual validation work. I personally feel pretty strongly we can’t cut corners, bang on lots of accelerometers on people and hope that something emerges from it. I think there needs to be mixed methods approaches, people actually doing observational studies with some kind of hard, ground truth measures about, say, functional dependence, possibly providing by clinical interview.

And some of the more blue sky areas that we might see, hopefully, some developments in, is using things like gait measures or movement sensor measures in community settings, to give us insights into symptoms that are really hard to investigate ecologically, or really in a valid way through pen and paper, or computerized tests. So the main one that comes to mind is spatial navigation. For example, the people I work with, with PCA, a core feature is environmental agnosia, but it’s not like there’s a very straightforward measure, at least that I’m familiar with, and I’m very happy if someone wants to suggest it to me, where you can just go off a three-point checklist of that environmental agnosia. So for us to really delineate, are there different profiles of topographical disorientation for some people? Is it more of a heading disorientation for other people? It’s more egocentric or allocentric? I think there might be some promise about sensor-based measures to really explore these symptoms but in everyday environments. So, are we still waiting for the ideal measure that will allow gait to become a mainstream element of diagnosis?

Dr Riona McArdle:

I think Silvia will probably have some very good insight into that, actually.

Dr Silvia Del-din:

I think so. So, everything that has been said so far, I think really points towards the fact that there is still a bit of work to do. More collaboration, more openness of methods, more maybe sharing of longitudinal databases, and more validation work to be done. When all these boxes maybe have been ticked, definitely it’s going to be, I think, the cream and the goal, that gait can become a mainstream element of diagnosis, if not a complementary, maybe, element of diagnosis because of the way that we know that clinical scale and clinician will always want to potentially use the clinical scale or the clinical-based assessment. But if we can get these analytics right, we’ve got data to validate and also to track, again, for example, you’re validating in a specific cohort, and then you can actually double check in a very similar cohort if what you’ve done is actually valid. And we can really create, I think, a very robust screening tool that can really help with diagnosis, support clinical decision-making, and as I say, be complementary also, in the clinical, if you want, field as well.

Dr Anna Volkmer:

Thank you very much. I think that’s all we’ve got time for today. It’s been such an exciting and passionate discussion. I’ve really enjoyed it. So, thank you to all our guests: Riona, Keir, and Silvia.

Now, we have profiles on all of today’s panellists on our website, including details of their Twitter accounts, if you would like to ask any follow-up questions, to follow them on Twitter and reach out.

And if you’re researching gait, or using gait yourself, we’d love to hear more. Drop us a tweet using the hashtag, ECRDementia, or add a comment to this post. Now, while I’ve got your attention, I’d like to remind you that we have a great website. It’s called dementiaresearcher.nihr.ac.uk. So we’d like you to register today, to get our weekly updates. And there you will find daily blogs, events, and details on all the latest funding calls. And regular blogs from myself, I should add. So, thank you for listening. Thank you so much for listening, and see you again.

Voice Over:

Brought to you by dementiaresearcher.nihr.ac.uk, in association with Alzheimer’s Research UK, and Alzheimer’s Society, supporting early career dementia researchers across the world.

END