 Now today's webinar, so advanced to the next slide, is the Global Importance of Frailty and Prefaility in Middle-Age Adults, a Pure Study. Let me introduce our speaker, Dr. Dale Leong. His current positions are as assistant professor in the Department of Medicine at McMaster University, a staff cardiologist at the Hamilton Health Sciences and an investigator in the Population Health Research Institute. Dr. Leong graduated from the University of Adelaide Medical School with Dean's Listing and Honors for Academic Excellence in 2000. He completed his cardiology training, Doctor of Philosophy, Master of Public Health, and Master of Biostatistics Degree at the University of Adelaide, Australia. He's completed a post-doctorate fellowship in cardiovascular imaging at the Leiden University Medical Center in the Netherlands before relocating to Canada. Dr. Leong has received a number of accolades, including the E.J. Morgan-Campbell Career Award at McMaster University 2014. He is a clinician scientist at the Heart and Stroke Foundation of Ontario. He's published over 100 articles and peer-reviewed journals, including the New England Journal of Medicine, Lancet, JAMA, Circulation, European Heart Journal, Journal of the American College of Cardiology, and Blood. So now we welcome Dr. Leong to begin his talk. So thank you very much everybody for your time today and for joining me at your lunchtime to hear this, hear our presentation and thoughts on this topic. So to start with, and please do interrupt me if you have any questions or anything I say is unclear or not coming across well technically or information wise, but is that slide, there seem to be blank. Can anyone see the slides? So I have what is frailty and one bullet point. There we go. Sorry, there's just a quite a lengthy lag between when I click next and when the slide appears, but there's the title slide that I see now. So hopefully everyone can see that. And so really the first question that I thought I would try to address is what is frailty? And I think that everybody has some conception of what frailty actually is. You have a mental image of it, but it can actually be defined by something like this definition here and there are numerous examples of these types of definitions available. We could consider frailty and aging associated susceptibility or vulnerability to poor outcomes when you're challenged by a physiological stressor. And so that makes a fairly good sense. The issue is really how can we measure this for the purposes of research? And I'll advance to the next slide, but it isn't appearing on my screen. Can you see the next slide at all? Here we are. It's just very slow. Apologize. It doesn't seem to be laggy, but we'll deal with it. We'll deal with it. So for the purposes of research, we can actually measure frailty in a number of different ways. And if you look through the literature, there are at least 25 different scoring systems for frailty. And I think the important message that tells us is that there is no perfect tool, there is no gold standard for evaluating whether an individual or a participant in a clinical study is actually frail. So with that in mind, what I thought I would do next is show you, when the slide appears, some of the approaches that we have adopted and that others have adopted as well. For the purposes of our research, we adopted two different approaches to measuring frailty. And I think it's fair to say these are probably the two most widely cited approaches. The first is an approach called a cumulative deficit approach or cumulative deficit index. And when you do this, what you do is you count for each individual in a study how many so-called deficits they have. And you express that as a proportion of the number of deficits that you've counted in total. So you might count at least 20 of these so-called deficits and ideally at least 30 to 40 of them. And I'll go into some more detail about what constitutes a deficit and what you can consider a deficit. But basically these are symptoms or physical signs that are abnormal, the presence of previous diseases or laboratory abnormalities. Because you're expressing these deficits as a proportion of the number of deficits counted, you then have a ratio or a fraction at the end of your count. And this fraction is what's known as your cumulative deficit index. And typically there are thresholds that have been described in the literature, whereby if you have less than one tenth of a point one of the counted deficits, you're considered non-frail. If you have more than 0.21 or some references more than 0.25 deficit index, so more than a quarter of the deficits that are counted, you are considered frail. And between those, you can be considered a state called pre-frail, which is between being non-frail and being frail. So that's one approach there. The other approach is what has been called a phenotypic approach, where you measure a number of characteristics that each participant has. So they're phenotype, if you like, and you can count them up as well. So for instance, with the classic phenotypic approach that was described by the degree almost 20 years ago, you measure muscle strength. And if less than the lowest quintile muscle strength that's considered to be low strength, you can measure physical activity in various ways, including a physical activity questionnaire. You can measure gait speed with a number of tests of walking speed. Typically, for instance, one might use a time to get up and go a test. And there are simple questions that you can ask to see if an individual has lost weight unexpectedly. And the sort of threshold that might be used would be, say, four kilograms weight loss unintentionally over the past 12 months or so. And lastly, we asked about exhaustion, which is very much a subjective complaint that an individual might have, whether or not someone has that. And so according to Fritz's initial description, if you had three or more of these characteristics of frailty, you were considered to be frail. And if you exhibited one or two of these characteristics, you were pre-frail, and none of the characteristics you were considered non-frail. I'm just waiting for the slide to tick over. I hope that's going to tick over. Is anyone able to see the next slide? They're actually coming through pretty well for me. Are they? Yeah, really not for me. So what I might actually do is if everyone's able to actually see these slides except me, it may be my internet connection. I'm going to open up the presentation separately so I can actually see what I presume that you're seeing as well. So the next slide that I have should be a slide asking the question, what constitutes a deficit in this cumulative deficit index? And so basically, as I said before, what I've listed here are the properties of something that you might want to consider as counting as a potential deficit in the cumulative deficit index that you're generating. A deficit should increase with age, but it should not saturate too early. And what we mean by that is we look at characteristics that tend to become more common as you get older, but we don't want them to be so common or develop so quickly that everyone by the time of say middle age has it. Secondly, you want these deficits to be associated with health. So for example, there isn't a very strong association between I don't know the presence of wrinkly skin and health outcomes, hypothetically, let's say. So even though wrinkles do increase with age, they're not specifically associated with health. And so they would not be an appropriate thing to count as a deficit as an example. Thirdly, we like the deficits to really cover in total a range of bodily systems. So you don't, for instance, want all your deficits to be focused on just the heart. You want to ensure that you gather information about a range of physiologic parameters for an individual. And lastly, most deficits are considered binary. So if you have them, then you get a score of one for that deficit. And if you don't have it, you don't get anything for that deficit. But some can obviously be weighted. As an example, you might consider muscle strength. I think we generally accept that the stronger you are, the greater your muscle strength, the better your physical condition. And so you might choose, if you're measuring hand-grip strength or muscle strength in participants, to divide up that one point, the lack of muscle strength, into a range. So for instance, if someone is in the top quintile of muscle strength, you might say that they have a score of one. If they're in the second quintile, a score of 0.75, the third quintile, a score of 0.5, and so forth there. So those are the sorts of ways that we approach trying to sort out what could be considered a deficit and how to use it in a cumulative deficit index. So on the next slide, I've actually shown some examples. As I've indicated on this slide, for the cumulative deficit index, I think at least 30 to 40 deficits are desirable. And the sorts of things that may be quite relevant to include in such an index include, for example, a past history of things like cancer or myocardial infarction, subjective feelings that the participants might express, like feeling depressed, feeling fatigued, and so forth. You can include functional parameters. So for instance, if they respond yes to needing help dressing or yes to difficulty on the stairs, that would be considered a deficit. And finally, you might consider actual physical measurements like group strength that you can take or blood pressure. And as I kind of alluded to before, what may not be appropriate for a cumulative deficit index is things like needing glasses, because we know that it's very common by the time you're middle-aged or older to need glasses. And so that really doesn't discriminate between someone who's more frail and someone who's less frail. And similarly, hair color is something that increases with age. It doesn't necessarily saturate too early, but it's not specifically associated with health and so is not so relevant for a cumulative deficit index. So hopefully that is something that's in clarity into how we construct these things. So on the next slide, what you've seen, what I've demonstrated is a group strength as an example of a measurement that might be included in the phenotypic index. So I'm actually turning my attention now to how we construct a free type of frailty index based on participants' physical measurements and characteristics. And the point I wanted to make with this slide is to address the question, how do you determine that someone has low group strength or low muscle strength? Because we've indicated in that freed index you might recall that low muscle strength is one of the five criteria that you use. And I wanted to make the point that this is not entirely clear. What people have traditionally done is taken folks with the lowest group strength for their sex and for their height and used that to determine whether they score a point on the phenotypic index of frailty or not. The challenge that we faced in our research was that as you can see from these graphs, in both men and women, muscle strength actually varies not just with age, as I said before, but also with where you come from and what your ethnicity is. So you can see that, for instance, folks from Europe and North America have significantly higher muscle strength for the same sex and age compared to folks from South Asia on the graph presented there. And so what we don't really know, and because global research into frailty, I think is something that's fair to say is in its infancy. We don't know whether we should use one global cutoff for individuals of a particular sex and age to determine what low muscle strength is or whether we should use ethnicity specific cutoffs, for example. So this is something that we need to do further research on and we haven't completely resolved. So while that might not exactly help you with research that you're doing, it is something that needs to be thought of and needs to be, I think, evaluated in a couple of different ways through sensitivity analysis. On the next slide, you see a similar principle that comes to physical activity. So this is data from the pure study that I'll describe in more detail later on that Scott Lear's published. And again, it shows that there is substantial heterogeneity amongst countries of different income as to what level of physical activity is considered normal. So you see that in high-income countries, some of us prize folks in general are more physically active than in lower-income countries and also the source of that physical activity varies. So recreational physical activity is seen almost exclusively in high-income countries, much less in lower-income countries. And so again, when you refer back to that free frail index and you ask yourself, is a particular participant low physical activity or not, the next question is what cutoff do we use for physical activity? Do we use a reference range that is really catered towards high-income countries or should we use something that's specific to where an individual comes from? And the truth is that, you know, levels of physical activity have not been studied as well in low-income countries as compared to high-income countries. And so those reference ranges, if you're going to use country specific reference ranges, are a little unclear. So these are some of the challenges that we currently face and that we need more research to try to clarify there. So having given you that kind of background about how we evaluate frailty and what some of the challenges are, I think the next question that's pretty intuitive is how common is it? And really, the answer to that question is that it depends. It depends as I kind of have been talking about, how you actually measure it. And it also depends what population you're looking at. So on this next slide here, we demonstrate some of the other previous work that's looked into the estimates of frailty prevalence. And there are a number more of these that I haven't presented. But I think you're going to get the idea that most of these studies come from older men and women from North America predominantly into a lesser extent Europe. The other thing to take away is that all of these studies evaluate frailty in different ways, in one of those 25 more indices that I mentioned at the beginning of my talk. What does come through though is that amongst older men and women from predominantly North America and Europe, it doesn't really matter how you measure frailty. Frailty and its precursor, if you like, pre-frailty are common. So if you look on the right hand side of this table, you'll see that a sort of Rockwoods scale there of three indicating someone who's quite frail is present in one in 20 adults aged over 65 in Canada. And Cawthon in the bottom row found similar finding in the U.S. Woods actually found a substantially higher prevalence of frailty using the freed phenotype, which obviously does differ from the cumulative deficit index. But I think it's based on the frailty and pre-frailty are certainly common in these data from older individuals from wealthy countries. And the next intuitive question is if we accept that frailty is common, does it actually matter? So I would argue that yes, frailty is very important. And I don't think that will come as a surprise to anyone. And here is some data to support that contention. So from those same papers I presented on the previous slide, you can see there that as frailty increases using the Rockwood cumulative deficit type of index, the risk of death increases quite steeply. And similar findings can be had from other studies, whether you measure frailty using a phenotypic type approach or another type of scale. The presence of frailty is associated with virtually a two-fold increase in the risk of death. And that's fairly consistent, irrespective of how you measure frailty. So given that background, and I think that that background may be somewhat familiar to most people who would be tuning in today. I think the real interesting questions are what do we consider to be the major knowledge gaps when it comes to frailty? And the specific knowledge gaps that I think exist when I read this sort of literature are as follows. Firstly, given the sort of relative paucity of information from global populations, are there actually differences in the rates of frailty amongst people from different parts of the world? Secondly, frailty has traditionally been thought of as a condition affecting older individuals. But to what extent does it actually affect people who are younger than that? Because that may actually be the seed. It may be the starting point for the development of later frailty as you get older. And finally, what I don't think has been explained very well is how frailty actually leads to premature death. We saw from the previous slide that it certainly seems to, but we don't know exactly how. It's not like a disease where you can die clearly of a heart attack. You can have a cancer that stops your vital organs from working. But we don't know how frailty actually leads to death. And so we'd like to try to address this in some of the research that we've been done. And we'll start by talking about the global prevalence of frailty. So there's a little bit of research done into this. As you can see here, this is some data pulled from a systematic review looking at the prevalence of frailty in middle-income countries according to the World Health Organization classification of country incomes. And it is limited to four countries outside of high-income countries. Again, as before, it's limited to all the populations in these countries. And they have sample sizes which are moderate, I think it's fair to say. What they show here is that frailty once again appears to be fairly common. What is challenging, though, is whether we can or whether we can compare these prevalence rates that we've indicated on this slide with the prevalence rates of frailty from high-income countries or indeed compare frailty prevalence amongst the countries on these slides. And the reason is that firstly, the approach to measuring frailty differed from between studies. And secondly, even though they included as a general rule older populations, if you actually look at the age distributions of these populations, they do differ from one another. And so it's very difficult to conclude that one country or one region has higher frailty genuinely than another country or another region and exactly what the prevalence is. Here's another piece of evidence that might help inform us. This is data from Europe where they mentioned frailty using a cumulative deficit type of approach in community-dwelling adults aged 50 and older. And they plotted the wealth of the country on the horizontal axis against the frailty index, that ratio I mentioned before, on the vertical axis. And they found that there was a genuine sort of inverse relationship whereby the higher the country income, the lower the prevalence of frailty, if you like. But once again, this data is actually limited to a fairly narrow bandwidth. Most European countries are really high or middle income with very little information on low middle and low-income countries. And also it's using an approach which does have some limitations that we can speak to. So I think that essentially the bottom line from my take on it is that we have somewhat limited information on how common frailty is globally and how it compares in different regions. Secondly, does frailty begin from middle age? So there have been a small number of studies that have tried to address this, but only really one that I can find that really looks at a younger population. And that is again a study by Ken Rockwood looking at adults in North America aged over the age of 20. They didn't really try and divide these participants up into frail versus non-frail. But what you can see from the frailty indices that were reported in this paper is that even in middle age 45 to 64, there's a suggestion that frailty may actually be reasonably common with a mean frailty index of 0.16, remembering that a frailty index of 0.21 or more might be considered to be frail. But aside from this data, there really is a very limited amount of information about how common frailty is in middle age and what it actually means if you're found to be frail in middle age. Lastly, how does frailty lead to premature death? That is something that I think there really is a substantial gap in knowledge about and something that we hope to be able to address in our research moving forward. So to get on to the pure study, which is really the title of my presentation today. For those of you who don't know, pure is a study of nearly 200,000 adults from 26 countries around the world. It was initiated by Dr. Selim Yusuf, who's my mentor here at McMaster University. And it now spans every inhabited continent except Australia, which is a source of frustration to us, but be it as it may. Pure recruited adults age 35 to 70, with a median age of 50 years, it turns out. And these folks were all living in the community. As it turns out, just the slight majority of these participants were women. And it is a study that's been being conducted for a long time now, so that the median follow-up across all all participants is approaching nine years. So this is a fairly substantial study. Amongst a number of measurements taken at baseline from pure participants was grip strength. We measured it using a hand-grip dynamometer. And you can see the image on the on the screen there. And what we found in previous work that we've done was that grip strength alone varied significantly between countries and obviously between sexes. And you can see the various pure countries at this stage. We only had 21 countries with hand-grip strength measured. Grip strength varies markedly between, let's say, countries in local areas and countries more high income like Canada and Sweden. But it was really this piece of research and what I'll present on the next couple of slides that piqued my interest in frailty as a phenomenon. So having measured grip strength at baseline and all these individuals, we examined the relationship between that measurement and the subsequent risk of death. And as you can see here that for every five kilogram reduction or lower grip strength that participants had, the risk of death increased by 16%. And that was consistent for both cardiovascular and non-cardiovascular death. And this tells us that grip strength may not be something related to any particular disease stream, but it might be something a little unrelated or independent of a particular disease. And we went on to look at the relationship between grip strength and your case fatality. And so what we mean by that is if during the course of those eight years or the time we did this study, six years of follow-up, you developed a disease, be it an MI, a stroke or a cancer, and you can see those displayed on the horizontal axis there, the grip strength where you just to divide up into high, middle, and lowest grip strengths. You can see the on the vertical axis, the case fatality rate, which is the risk of death after the development of any of these conditions. And what we're showing is that no matter what disease you've got, whether it be an MI or cancer or pneumonia, if you had low grip strength in the blue bars there, your chances of dying were substantially higher than if you had even middle or high grip strength. This tells us that grip strength seems to be very much a protective factor or muscle strength, I should say, as a protective factor against dying from some sort of insult or disease. And that's a nice thing because that actually ties back in with the original definition of frailty I mentioned before on the very first slide where it looks as if muscle strength is telling us exactly what we hope it would when we think that it's a marker of frailty. So since doing that initial research on hand grip strength and developing an interest in frailty, we then have gone on to measure frailty in the pure cohort in two ways. One is using a cumulative deficit index using 47 characteristics that are measured in pure, and we use those thresholds that I mentioned before considering someone pre-frail or frail. We also measured frailty using a phenotypic approach. In pure, we have only measured grip strength, physical activity using the international physical activity questionnaire, and there was a simple question on whether there was unintentional weight loss of more than three kilograms in six months. And so we considered someone to be frail if they had two or more of those three characteristics and pre-frail if they exhibited one. So shown on this slide here are the characteristics of the pure participants divided into whether they were non-frail, pre-frail, or frail and how we measured frailty by phenotype or cumulative deficit index. What you can see is that the proportion of individuals who are non-frail, pre-frail, and frail are reasonably similar irrespective of how we measured frailty. Similarly, there wasn't much discrepancy between men and women, and it didn't seem to matter too much how you measured frailty. Perhaps women were slightly more frail when evaluated with a cumulative deficit index. What we also find is that as one might expect, frailty increases with age, so that's nice because it tells us that there is some face validity to what we're measuring. And what was quite interesting is that frailty is inversely related with level of education. And as we all know, education is an important marker of socioeconomic status as well. So that if we consider the frail phenotype, for instance, the higher the education achieved, the lower the prevalence rate of frailty. It's sort of halves if you go from folks with only primary school education down to secondary school and even less for those who achieved university or college education. Similarly, with the cumulative deficit index and the bottom right cell, you can see that there was more than a halving of the prevalence of frailty as folks became more educated there. So it seems to be an important factor in what leads to the development of frailty. What we also find is that there is a complex relationship between alcohol and tobacco use and the occurrence of frailty. And I think it's fair to say that no very strong or clear pattern emerges. What is always complicated when measuring alcohol and tobacco use in these sorts of research is that the group that were formerly drinkers or formerly smokers have often quit smoking or drinking for health reasons. And those health reasons might play into why they might be more frail than people who are current drinkers or smokers, for example. So these sorts of data, while interesting, can be a little challenging to interpret. But what we show on the next slide I think is really one of the key findings from our point of view. And that is that it doesn't really matter how you measure frailty again. As you move from high income countries on the left side of each of these bar graphs to low income on the right side of each bar graph, the prevalence of frailty adjusted for age and sex of the population increases and increases fairly substantially. And I think the great strength of PURE is that because we measured frailty in a very consistent manner across these different populations, we can be very confident in these conclusions, whereas other trying to sort of extrapolate from previous work that only measured one population and comparing that with a different study using different methodology is a little more limited. What we show on this slide here is the relationship between frailty and the risk of death after five years. But what we've also done, understanding that the cumulative deficit index is in some respects a measure of multi-mobility because it includes in it many diseases. We've measured frailty here on that sort of access going into the slide using the phenotypic approach. And we've subdivided that into whether or not the individual had one or more or no diseases at baseline. So we're looking here at the relationship between frailty, level of multi-mobility and risk of death at five years. I think what the skyscrapers show us is that frailty and multi-mobility seem to have an additive effect so that for every level of let's say frailty, if you have more diseases, your risk of dying is high or put the other way for any level or any given level of disease, the more likely you are to die. In this Kaplan-Meier curve here, what we present is the risk of death in individuals who had no diseases at baseline that they knew of and who did not develop any disease like myocardial infarction or cancer or pneumonia during the follow-up. And what was really interesting is that again it doesn't matter how you measure frailty and this is a recurring theme. It doesn't matter how you measure frailty but the more frail you are, the more the risk of death is irrespective of whether or not you seem to develop or have any disease. And so this is a really curious funding because it does beg the question how does one actually die because of frailty. Here in this analysis we did something similar, we excluded those with any chronic disease and we looked at whether or not being frail and pre-frail predicted death as well as cardiovascular and non-cardiovascular death after adjustment for a range of confounders and those confounders you can see are listed there. And what you can see is the relationship is still pretty strong. If you are frail, even accounting for differences in age, sex, education and so forth as you can see, frailty is still associated with nearly a two-fold increase in the risk of death as compared to someone who is non-frail and that seems to be independent of whether or not you have a non-chronic illness. What we also found which is quite interesting and almost puzzling to us is that the frailty phenotype was not strongly associated with the risk of developing disease. So on the left hand column you see diseases whose incidents were measured during those eight or nine years or well eight or nine years of follow-up. And you can see that there really was not a strong relationship between being frail at baseline and actually contracting one of these diseases with some exceptions. So there is, for instance, a trend towards a positive association between being frail and developing heart failure, pneumonia or COPD. These didn't reach statistical significance, however. But interestingly there was no real strong association between being frail and developing marcal infarction or being frail and developing cancer. And in some respects that is an encouraging finding because there is some face validity to it. I can't imagine a mechanism by which being frail in and of itself would predispose us to developing cancer. Whereas if let's say we had found a positive association between cancer and frailty, one might be concerned that there could be reverse causation where it was the cancer causing the frailty. So while that is reassuring in some respects, what it still doesn't really tell us is how frailty causes death if it doesn't in fact cause a whole range of diseases. What we did find impure though is, as we presented for the group strength, being frail increases your risk of dying if you develop a disease. So again we have those incident conditions listed on the left of the slide and shown here are the case fatality rates depending on whether you are non-frail, pre-frail or frail. And what is very apparent as with the graph I showed before, the bar graph, is that the more frail you are the higher your risk of dying should you develop some sort of intercurrent illness or disease. So basically to try and summarize what our findings are shown here in this slide is what I think is the classic model of death and that is we all start off let's say healthy. During the course of our lifespan we will have or develop various risk factors for disease and they may eventually lead to the development of disease. If you develop disease you can either die from it or you can survive and it may leave you disabled but that surviving individual is always going to be at higher risk for future disease and so feeding back into this sort of cycle which eventually leads to death and that's how we were taught at medical school to think of how one develops disease and dies. But I think that our research shows frailty maybe have a very important role in this in this model in that if you develop disease frailty increases your risk of death there and so it seems to be a facilitator of death if you have a disease. But also we found that individuals who are apparently frail and have no disease that seem to develop a disease during follow-up do seem to be at increased risk of death and I think this is a source of much interest to us and we need to understand this population better. So I'm just going to finish up with a last couple of slides to speak to where we're heading with all of this. I think when we if we acknowledge that frailty is an important phenomenon then we next need to ask ourselves why do people become frail and I think there are a number of players that might feed into why people become frail and we can classify these as to some extent biologic for instance you know time itself is likely to contribute to frailty but so too might diseases and multimorbidity and then there are a range of modifiable determinants like your behavioral risk factors and non-modifiable ones like you know whether you're a man or a woman for instance and so you might actually subdivide those categories of potential determinants of frailty even further and people have done lots of research into these different categories of determinants for example there is a clear association between shortening of the telomeres and the chromosomes and biologic aging relationships between mitochondrial DNA mutations and biologic aging in the next column you see you know I think that there is almost by definition a relationship between a physical inactivity and frailty and so forth so these this is how we can conceptualize potential determinants of frailty what we're currently doing to try and get to what are the most important causes of frailty is we are in the process of measuring frailty amongst the pure participants on the left hand side in those who survive 10 years and what we want to know is if you are robust or non-frail at the beginning of this pure study and you then subsequently become frail what has led to that? Incidentally I should mention we have a similar dataset that we're developing in prostate cancer patients as well and so we consider frailty as I kind of alluded to before as you know a number of potential contributors to it and what we are doing is looking for potential biomarkers of frailty and these can be categorized in a number of different ways and I've listed one here that's occurred to us as we've done a scoping review of literature on the biomarkers of frailty and again there are you know many dozens of these listed in the literature but what we eventually hope to do is use advances in biomarker analytic techniques called multiplex analyses whereby we can analyze hundreds of these biomarkers with very small quantities of biospecimen or plasma to then identify patterns of biomarkers rather than what is traditionally done identifying single biomarker associations at a time and we feel that these patterns of biomarker associations kind of using the categories I showed in the previous slide may actually be more relevant in predicting who's going to become frail than any one biomarker by itself obviously this is a big undertaking we are measuring as I mentioned hundreds of biomarkers and they can be categorized in different ways and they will have relationships with the development of frailty which needs to be adjusted for by a number of confounders and so we are starting to develop these very very complicated biomedical models to describe how frailty develops and I think that these sorts of analytics can very easily and rapidly get beyond the brain capacity of poor folks like myself and so I think that we are going to need to leverage things like artificial intelligence to enable us to understand these complex relationships as we move forward and as our research keeps up to date with the available technology for doing the research so really to wrap up and allow some time for a discussion I think it's fair to say that both frailty and pre-frailty are common I think we've shown that they are more common in poorer settings and in lower income countries it's very clear that frailty can be identified from middle age and has prognostic significance from pure even from middle age and it's also clear that frailty leads to death by increasing susceptibility should you develop an illness but also seems to potentially play a role in leading to death independent of disease and I think we need more research to understand why that observation it might be the case and so lastly I think that the take home message for me is that reducing frailty given how common it is and how important it is may really be an underexploded way of reducing mortality and I haven't spoken this likely disability that can complement our existing efforts to prevent and treat disease and so on that note thank you very much for your attention thank you that was really a great presentation um I'll go ahead and open it up to questions just a reminder muting remains on that you can enter your questions into the chat box at the bottom right hand corner of the web next window at any time and we'll read through your questions so um Nicole I remember asked curious if frailty could be an indicator of recovery time and quality of the recovery so um would you like to speak to that so recovery I might have missed part of it was that recovery from was it recovery on quality of the recovery yeah okay I'm gonna take it the question is curious if frailty I'm sorry Dr the young can hear me sorry yes I can hear you yes I'm gonna take it that you meant recovery from let's say an illness or even an injury and I think the answer is almost certainly yes I think that the less frail you know the more frail you are the less likely you are to recover um and the slower you are going to be to recover uh from any major illness or injury I think we certainly see this clinically and anecdotally and you know this is something that could certainly be measured in research and and almost certainly has but I think that that conclusion is almost certainly correct to speaking speaking to the definition of frailty or the resilience or ability to recover from illness injury exactly so the vicissitudes of exactly so by definition frailty is a characteristic that enables recovery from illness or injury really the more pertinent question is to what extent are our tools and methods for measuring frailty able to detect that and if you are able to show that these sorts of tools are predictive of recovery time then again you have added face validity to how you measure frailty certainly well while we wait for a couple more questions again and I'll go ahead and pose one um so you know you you showed fairly well that that frailty does differ across countries it seems to be an important indicator of frailty did you look for groups across countries did you did you stratify based on on income country to country and look at that yes so you can divide stuff in a number of ways you can divide it up by country income which is what we did it you can divide it up by also by things like level education so irrespective of where you come from be it a high or a low income country the more educated that you are there is you know a generally pretty good relationship or correlation between education status and social economic status and so when you do that even cutting across countries there's again a very positive association in between level of frailty and education so clearly there is something about a level of wealth be it in a given community or be it in individual level wealth that seems to protect you if you are wealthy from a wealthy community from developing frailty exactly what that is i'm not entirely sure and i think it's likely to be multifactorial and we need to do more research and and we are doing this research to understand why you know you might hypothesize for instance that it is things like physical activity if you are wealthier or from wealthier communities you're more likely to be able to undertake recreational physical activity play sports it may be to do with diet quality so for instance you can afford to eat fresh fruit and vegetables which are notoriously expensive in lower income settings it may be with more education that you know that smoking is a bad thing and so you don't do it and you don't develop the risk factors and diseases that can lead to you becoming frail so i think it's likely that these sorts of wealth related questions about what is it about being wealthy that protects you from disease and frailty i think they're likely to be complex and multifactorial answers i think that these relationships are very strong and very true but we need some more sophisticated analytic approaches to understand exactly how well i would take the opportunity to do is make the point that it speaks very much to social equity and i think that if we want to improve health outcomes across communities in canada or wherever we happen to live it is important to bear in mind that the more equitable the distribution of wealth in the country the more likely it is that you have a homogeneously good level of health and in contrast if you have a society or community where there is a marked disparity in people's level of wealth you're likely to equally see a marked disparity in their level of frailty and in their level of diseases and illness and i think we can all think of examples of countries where that exists and so i think that this is important for health researchers and global health researchers to bear that in mind and try and advertise this very prominently and you that's probably speaking also to how frailty may be a facilitator of death and not not just a predictor of diseases that lead to death i i think so yeah absolutely that'd be fair to say yeah very interesting all right another question uh what preventive methods are best for reducing frailty so you're talking about blood tests for frailty you're talking about predictive models for frailty if as a health policymaker or a doctor that you say someone's frail or is at risk of being frail what would what could be rolled out to actually reduce yeah good question uh downside effects of frailty so there's been quite a lot of research done in typically short duration trials like about three to six months looking at exercise and diet and those are fairly obvious candidates and it doesn't really come as a surprise that if you can exercise intensively enough and you supplement especially if you have a nutrient deficient diet with things like a high quality protein that you increase muscle strength in that short term and you reduce markers of frailty the challenge really from a public health perspective is how you implement that and to give you some insight into the complexity if we talk about say diet i think again most of us will know what constitutes a healthy diet there is a lot of pushback or resistance to necessarily implementing that so if you consider fresh vegetables being a part of diet to get them to remote and rural communities requires refrigeration to get them to keep them fresh it requires a lot of cost because they are more expensive than foods that have a very long shelf life because they're full of preservatives and salt so there are substantial difficulties in providing that plus there are interest groups so there may be groups and lobbyists within the community whose interests are not to necessarily promote fresh fruit and vegetables and high quality protein but whose interests may be more to promote food which quite frankly is of poorer quality and so when you put all of these ingredients into the mix it does become very difficult for governments to legislate for healthier food choices and practices and to institute things like taxes or tax rebates to encourage that because there are lots of different facets to these sorts of arguments now with exercise it's a slightly different story because i don't think anyone is going to debate that exercise is a good thing and there is not i would imagine anyone who would be promoting a sedentary or unhealthy exercise lifestyle but the question really comes down to one of time accessibility and adherence just to give you an example i've heard that in sweden they actually the government provides tax relief for companies that allow their workers an hour a day where they are supposed to do dedicated exercise and i put myself that is a phenomenal public health policy and probably speak to the white swedes have you know such a high level of health across their country and so i think innovative approaches and solutions to things like physical activity like what they do in scandinavia is something that we could consider implementing if there's enough will and that would likely help with preventing or reducing frailty certainly well let's do three quick questions and then say goodbye i'm wondering if you adjusted for health care infrastructure public versus private health care access across different countries and whether there were any differences in prevalence of frailty and their associations with death so health utilization or care across countries a great question we to some extent that is incorporated into you know when we adjust for things like education and country income but as you rightly point out it's not completely it doesn't completely account for differences in health insurance and coverage these are data that have been collected in pure at you know a sort of a community level and it's certainly something that we could look into and that was from lily robert mine as part of a cumulative disease and do you not have complete confounding when you're analyzing rates of disease leading to death so maybe a little methods explanation for it yes so we've done a number of different iterations and i left this out for the purposes of keeping things simple we've done a number of iterations of the cumulative deficit index one that includes that it is like you know a history of cardiovascular disease a history of cancer and one that does not include a panel we've also done analyses whereby we exclude individuals who have any any chronic disease at baseline and actually irrespective of how you cut up the pie the findings are still the same and the effect sizes do not change very much from a risk of death of about two with frailty so you wouldn't necessarily think it but that's what we find okay and we have a request to review your biomarkers one last time the slides you mean yeah if i go back to this one here i think so just maybe give a quick brief overview of of your biomarker set yes this is not actually necessarily our final list we're currently doing a scoping review we're about a third of the way through to really get a bird's eye view of what people have done in the past and linked with frailty and again like i said the traditional approach has been to measure one of these markers or maybe two or three and compare it with some sort of marker of frailty usually in a cross-sectional manner the problem is that you know biologists are complicated and when you measure just one two or three markers you don't really capture the complexity of what a human a human being is so given these new sort of o-link platforms that we have available to us at phri where using just a few microliters plasma you can measure i think up to about almost a thousand biomarkers simultaneously we now have the capacity to actually measure a vast array of biomarkers and we're in the process of tabulating what list we actually want now that these that the platforms that you use are customizable to some extent but probably not to the extent that we would like having said that i think that we'll be able to come to a nice list the panel of biomarkers that gives us a good idea about the physiologic changes that are occurring that eventually lead to the development of frailty so this is kind of work in progress that we're presenting here certainly yeah that's fascinating well thank you very much dr now for such a really really great talk i enjoyed it and i think our audience did as well we appreciate your participation in the clc webinar series our problems thanks for having me thank you so i'd like to remind everyone that the clsa data access request applications are ongoing the next deadline for applications in september 24 2018 please visit the clsa website under data access to review available data for the information in details about the application process oh we have the survey back online it was about to say due to technical difficulties that we sent out but um it looks like it is up and running online remind everyone to please say located in the polling option we would appreciate your help with uh having our surveys delivered back to us either now or after uh we send it anonymously if you have any questions or concerns that we can help with about that write to us in the chat box and we can help um and also please remember that clsa promotes this webinar series using the hashtag clsa webinar and we invite you to follow us on twitter finally today is the first of our 2018 2019 monthly webinar series and we will have our next webinar in october uh talking about the canadian urban environmental health research consortium our canoe and its data linkage with the clsa we have uh three speakers we'll be welcoming jeff brock the canoe scientific director elinor satin the canoe managing director and danie doron canoe dating linkage specialist to speak so please go to our clsa website to register for our webinar webinar series soon and join us for next month's webinar thank you everyone for attending today's presentation