 Now on to today's event, which is again entitled, Exploring the Patterns and Impacts of Diet and Nutrition Among Older Adults in the CLSA, presented by Dr. Jacqueline Hurley and Dr. Rachel Murphy. So I'm gonna do a quick bio of each of our presenters. First, Dr. Rachel Murphy is a scientist in cancer control research at BC Cancer and an assistant professor in the School of Population and Public Health at UBC. She holds a PhD in Nutrition and Metabolism and completed her post-doc training and epidemiology at the National Institutes of Health. Her research program has focused on the role of nutrition in the prevention and control of cancer and the etiology underlying lifestyle, cancer and relationships. Dr. Jacqueline Hurley is an assistant professor in the School of Kinesiology and Health Sciences at York U. Her area of expertise is musculoskeletal biomechanics where her research interests include investigating mechanisms of musculoskeletal injury and developing effective exercise rehabilitation strategies for chronic conditions that commonly accompany age, including osteoarthritis and rotator pathologies. Now I will pass it on to our presenters and I believe Rachel's going first or I could have that wrong. Dr. Hurley's presented. All right, great. All right. Let's share my screen. Okay, so thank you very much. I'm very grateful to have the opportunity to speak in this webinar series today and share some of our CLSA research, investigating diet and nutrition risk and their relationship with mobility and general health in older adults with osteoarthritis. So I wanted to begin by acknowledging my wonderful co-authors and collaborators on this project beginning with Dr. Monica Malley. So Dr. Monica Malley is an associate professor at the University of Waterloo and extensively studies musculoskeletal biomechanics and osteoarthritis. And so the work that I'm discussing today was work that I conducted while I was a postdoctoral fellow in Dr. Malley's lab. So I'm very excited to be able to share this exciting work that we researched during my time at McMaster as well as at the University of Waterloo. I would also like to acknowledge Emily Wivinga who is Dr. Malley's research coordinator and project collaborator as well as Dr. Heather Keller who is a professor at the University of Waterloo and an expert in nutrition and aging. I also wanted to note that the research I'm presenting today, so our background or methods or results discussion as well as our tables and figures have been published in the journals of gerontology medical sciences. So the full citation could be listed there. So I wanted to start off by briefly talking about osteoarthritis. So osteoarthritis or OA is the leading cause of chronic pain and lower limb disability and it affects the knee most commonly. It is a degenerative disease affecting internal joint structures notably cartilage and it impairs work productivity and longevity. It also increases the risk for a variety of different chronic health conditions including cardiovascular disease and depression. OA is also extremely prevalent so it affects one in eight Canadians and does increase considerably with age. It is hiring women, particularly over the age of 50. Now this number is expected to dramatically rise if not double in the coming decades. Now surgery is the most successful treatment but the resources showing below have reported that greater than 190,000 patients were eligible and willing for surgery but didn't receive it. And this will probably be speculating but I suspect with COVID it would probably be even more and this number is expected to rise if not double within the next 30 years. So therefore it's critical to determine some effective non-surgical interventions that allow us to reduce the painful symptoms and functional deficits in osteoarthritis. It's also expected that one in five Canadians will be overweight or obese by 2024 and a relationship does exist between obesity and OA. And so the literature has shown that overweight and obese individuals are actually at 2.5 to 4.6 times greater risk of having the osteoarthritis. And so this data was from the meta-analysis of a systematic review that showed that the risk of OA is increased 35% with a five kilogram per meter squared increase in body mass index. And so this risk of OA related to obesity can be attributed to several different factors such as higher mechanical loads. So a greater body mass would cause you to have greater loading on those vulnerable compartments of the knee as well as physical inactivity. So this can go both ways as physical inactivity can elevate our risk for OA by a reduced muscle mass or poor muscle quality but also a painful OA joint may reduce our ability or willingness to exercise. There's also different pro-inflammatory processes which have been associated with increasing pain and worsening function. And so there are several different interventions such as diet and exercise interventions which may yield improvements in both obesity and OA related symptoms. So a randomized controlled trial that investigated both diet and exercise among overweight or obese individuals with OA actually showed that by reducing body mass by 10%, this was related to improvements in physical function. And so this was by Dr. Messian and colleagues. But Dr. Malley, among others have conducted several research studies focusing on exercise for knee osteoarthritis and have demonstrated these improvements in pain and self-reported function, mobility performance and strength. And so we also started to become interested in studying different aspects of nutrition and how that might affect OA related symptoms. A specific food intake may affect OA with again research showing that better quality diet may be associated with better mobility performance. In this case, mobility performance was measured using a chair rise test. And then dietary fiber was also related to a lower risk of symptomatic knee OA. And the suggested pathway here for this relationship was the effective fiber on reducing body mass index which would subsequently then affect those inflammatory processes that are related to disease development and the risk of pain. And so a lot of the diet trials in OA are typically very strict. So the caloric intake wouldn't necessarily reflect one that we might consume regularly. So we started to think about nutrition risk as nutrition encompasses attributes other than the physical food being consumed. And so nutrition risk screening allows us to examine the behaviors related to nutrition and non-nutrition. And nutrition risk wasn't evaluated in OA. So the purpose of this research was to examine whether aspects of diet and nutrition risk relate to physical capacity and general health in older adults with osteoarthritis. And so we hypothesized that consuming more high calorie snacks, consuming lower dietary fiber as well as being at higher nutrition risk would be associated with poor mobility, grip strength and self-reported general health. And so for this, we used data from the Canadian Longitudinal Study on Aging. And so we specifically use baseline data from CLSA participants recruited for a collection at the comprehensive database. Both qualitative and quantitative measures were captured in adults 45 to 85 years. And the data for analysis for the sample was collected between May 2012 and 2015 through both face-to-face at home interviews and at data collection sites. We also had a number of inclusion and exclusion criteria. So for the inclusion criteria, we included participants that were between 45 to 85 years at baseline. And they would need to self-report a diagnosis of hand, hip, or knee osteoarthritis by a physician. And so the question in particular was, has a doctor ever told you that you have osteoarthritis in the hand, hip, and or in the knee? We also had a number of different exclusion criteria that reported there. I won't read them all out, but essentially we were looking more at neurological and respiratory conditions, as well as an incomplete data set. And so any participant that had missing data for any of our independent or dependent variables were excluded from analysis. So starting with our independent measures. So we evaluated certain aspects of diet using select questions from the short diet questionnaire. And so this was classified with the variable NUT or NUT. And so specifically we were looking at question one, which asked about high-fiber cereal intake. And then questions 25 to 28 that looked at high-calorie snacks. And so there wasn't a specific question asking about high-calorie snack intake, but instead question 25 looked at ice cream, ice milk, frozen yogurt, milk-based desserts. Question 26 was more salty snacks, so chips and crackers. 27 was cakes, pies, donuts, pastries, cookies, and muffins. And then 28 was chocolate bars. And so what we did was they were asked to record the number or the number of servings and then the unit of measure. So per day, per week, per month, per year. And so we combined the high-calorie snacks into one single independent variable, labeled our NUT HC or high-calorie. And then high-fiber cereal was a separate independent variable and again classified as NUT FDR or NUT fiber. We also evaluated nutrition risk. And so this was evaluated using the modified screen to abbreviated tool. And so again, this is one of the questionnaires within the CLSA database. And this was developed by Dr. Heather Keller, who is one of our co-authors. And it is an 11-item questionnaire that asks participants about several different behaviors related to nutrition. And so specifically, we'll ask about changes of weight. So compared to six months ago, it would ask if they gained weight, lost weight, or stayed the same. It would also ask about how much weight they lost or gained in the six-month period. It asked questions pertaining to appetite. So how often they generally skip to meals, how they would describe their appetite. There were also questions pertaining to physical challenges while eating. And so how often they choke cough or have pain when swallowing food or fluid. And then there was also a question about or a couple of questions pertaining to diet. And so specifically asking about fruit and vegetable consumption as well as fluid consumption. And then lastly, there was some questions related to meal patterns and socialization. So how often you eat a meal each day with somebody else? Do you cook your own meals? And then asking them to comment on certain statements related to meal preparation, such as enjoying cooking or sometimes finding a chore, as well as whether or not they're satisfied or unsatisfied with the quality of food prepared by others. And so the scoring for this, participants who had a score greater than or equal to 43 would be deemed lower risk. Moderate risk was between 38 and 43 and higher risk less than 38. And so I've included the citation here for some of the psychometric properties for this screen two tool in the bottom right corner. In terms of our dependent measures, we first evaluated physical capacity. So physical capacity was evaluated using grip strength and mobility. And so grip strength was captured again through the CLSA using the tracker freedom wireless grip dynamometer in a figure from the website is shown here on the right. And we chose to evaluate the average of three repeated trials. For mobility, we used a pooled index of four different mobility tests, including a four meter walk. So this was the total time to walk a distance of four meters, a one-legged standing balance is the best attain time for standing on one leg up to a maximum of 60 seconds. A chair rise test, so the average time to rise out of a chair five times, and then a timed up and go, which is the time to rise out of a chair, walk three meters and then return to the chair to the seated position. And so instead of looking at these four different mobility tests independently, we chose to create a mobility index where a higher mobility index equal better mobility. And in both cases, a higher grip strength and better mobility would represent better physical capacity. And we also evaluated general health. And so we evaluated the total score of three questions pertaining to general health with participants being asked to rate their general health, their mental health, as well as their own healthy aging as excellent, very good, good, fair or poor. And again, the higher score here would represent better general health. We also chose to include a covariate center analysis that we believe might impact physical capacity in general health in OA based on previous literature. And I have a list of those here. One that we did include as well at the bottom was OA type where there were seven different classifications. So in this case, we looked at those who had hand OA only, hip OA only, knee OA only, and then the combination of the two and the combination of the three. For our analysis, we conducted several linear regression models to study the relationship between independent variables and covariates with each of our dependent variables. And so we first analyzed a covariate model where we examined all the covariates that were described in the earlier slide with each of our three dependent variables. We then conducted separate models for each of our independent variables, again, after accounting for those covariates and ran a likelihood ratio test to determine whether those models were statistically different from that covariate only model. And so one thing I just wanted to point out on this slide is you could see that I have four different independent variables. We have our nut HC, which is our high calorie snack intake, our nut fiber, which is our high fiber cereal intake. But then we chose to look at the nutrition risk score, as well as the nutrition risk classification where they're grouped into those at low risk, moderate risk and high risk based on those classifications that I described earlier. Now looking at some of our results and starting with participants. So from our initial sample of approximately 30,000 participants, 7,900 roughly met the inclusion criteria and 3,700 after the exclusion criteria. Now our final sample size used for analysis was 1,404 because those are the participants from this sample that had a complete data set and some information pertaining to the participants can be seen here. Now starting with grip strength. So our covariate model was significant with older age, female sex, greater depressive symptoms and lower income related to lower grip strength. But for grip strength, neither of our dietary variables nor the nutrition risk variables were shown to be significant. For mobility, again, our covariate model was significant, but here our nut fiber as well as the nutrition risk score were shown to be significant with greater fiber intake and lower nutrition risk associated with greater mobility. So again, as hypothesized, but neither the high calorie snack intake or the nutrition risk classification was significant. Now there was a moderate effect size despite not being significant between our low and our high nutrition risk classification groups with higher nutrition risk classification again associated with lower mobility than that lower nutrition risk classification group. And then lastly, in terms of general health, again, the covariate model was significant. Neither of the dietary variables were significant, but our nutrition risk score was significant again with higher nutrition risk associated with lower general health. This figure here shows again our nutrition risk classification which again was significant where we have general health index on the y-axis. Our nutrition risk classification on the x-axis with the green representing a low nutrition risk, moderate in that sort of yellowy orange color and then high nutrition risk in red. Above that, you'll see the statistical differences as well as the effect size. And you see there's significant differences between low and moderate nutrition risk and low and high nutrition risk where higher nutrition risk was associated with lower general health. And so overall nutrition risk was significantly associated with both mobility and general health in hip, knee and or hand osteoarthritis. And these findings do support previous literature that's examined the quality of life among frail seniors that have shown that higher nutrition risk has reduced our self-reported physical health in this population and it's been associated with functional decline. Also the intake of examined dietary items were not significantly associated generally with physical capacity or general health with the exception of higher high fiber cereal which was related to better mobility. And again, this supports previous literature that showed that fiber can reduce adiposity and inflammation which again are associated with pain in the leg. And then certain covariates and comorbidities explained a large amount of variants in both physical capacity and general health. I did also wanna mention a few study limitations. So first, we did not provide a comprehensive diet analysis but rather chose to focus on two areas including high fiber cereal and high calorie snacks but certainly there's other aspects of diet that could have attributed to both physical capacity and general health. Also you may note there was several participants excluded due to missing data but I wanted to note here that we did run separate samples for each of our dependent measures. So where they would have different sample sizes among the three different dependent measures and there were no changes to the results. And then lastly, there were other potential factors that could explain variants in physical capacity and general health. For example, other diseases or conditions that would be linked to poor health outcomes among aging adults with osteoarthritis. And so the primary take home message of this work was that nutrition risk is important for older adults with osteoarthritis where it demonstrated that these behaviors surrounding nutrition are important contributors to both mobility and general health in OA. And so these results offer some new potential suggestions for conservative nutrition-based interventions to improve both physical capacity and health among older adults with OA. So I wanted to thank everybody for their attention today again for being invited to speak here today and again to my co-authors as well as Dr. Stratford and Dr. Gattie for providing statistical support as well as our research support. So the Schlagel University of Waterloo Research Institute for Aging where Dr. Keller is a research chair in nutrition and aging as well as NSERC for the Discovery Program for awarding a grant to Dr. Malley to support operating costs and then CIHR again for providing postdoctoral support for me during my postdoc. And then again obviously to the CLSA. So thank you very much. And I think questions are gonna be deferred to the end so I can pass it along to Dr. Murphy to share her work. And I will stop sharing my slides here. Thank you so much. And yes, so we will pass the torch on to Dr. Murphy. Okay, great. Thank you very much also for the invitation to speak today. I have a long-standing interest in healthy aging and disease prevention, although more recently I've been focused on cancer prevention. I actually completed my fellowship at the National Institute on Aging in the US and one of the things that attracted me to that fellowship at the time was the ability to work with some of these leading epidemiologic aging studies in the US. So when I first learned about CLSA was when I was starting to move back to Canada after my postdoc and I was really thrilled that there was gonna be a resource like this within Canada. And the work that I'm presenting to you today is actually some of the first that I undertook when I started my first position at UBC. So it won't spend much time on the rationale for studying health aging given the audience today, but very briefly over the past 100 years life expectancy in Canada has continued to increase in men and women, although it is flattening a little bit in more recent years. So the average life expectancy is around 82 years with some variation, especially provincially. So for example, in BC where I'm based there's one of the highest life expectancies of the Canadian provinces. However, the increases in life expectancy have not been equaled by increases in disability free life expectancy. So this table shows findings from the US. So the numbers differ slightly from Canadian figures on the prior slide, but overall you can see that the changes in life expectancy are not equaled by the changes in disability free life expectancy. So for example, in men, if you can see my cursor here and the changes in life expectancy is about 9.2 years and the changes in disability free life expectancy is about half of that. So around four and a half years. So what is the role of diet in life expectancy or in disability free life expectancy? There are of course many nutritional requirements for growth and development as well as for maintenance of overall health. So things like the musculoskeletal system, but diet also plays a very large role in many of the most chronic, common chronic diseases in Canada. So for example, about 40% of cancers can be prevented through a healthy lifestyle, which includes consuming a healthy diet, maintaining healthy body weight, being physically active and minimizing alcohol consumption. Similarly, it also plays a large role in the prevention of other chronic diseases such as stroke, heart disease and diabetes. So diet is also intrinsically linked to other health behaviors that are important for overall longevity and healthy aging. So for example, body weight, alcohol consumption, physical activity, smoking and potentially stress and sleep as well. This is an infographic from the compare study. So this is just to provide some high level context on the importance of diet for overall chronic disease prevention and in this case, cancer. So compare is a Canada wide study which aims to estimate the current and future burden of cancer due to modifiable lifestyle environmental and infectious disease or infectious risk factors, I should say. So this still shows the number of cancers that can be prevented in Canada in a given year. So notably about 6,700 cases are attributable to low fruit consumption, 3,500 to low vegetable consumption. And then over here about 1,700 are attributable to red meat consumption. So despite the importance of diet, we know on a population level, Canadians largely don't have diets that align with overall recommendations for health such as the dietary guidelines, as well as for chronic disease prevention. So just some very brief statistics here, as I know many of you already know this, but about 50% of women and 70% of men have energy intakes that exceed their needs, but 40% of women and 50% of men don't meet their fruits and vegetable, the daily recommended intakes, in this case around five servings per day and about one in four half intakes are above the recommended range. So what do we know about the relationship between diet and longevity? One of the most widely studied populations with respect to diet is in Okinawa, Japan. Perhaps you've heard of the Okinawa diet, I feel like it's been on the bookshelves and a lot of those popular dietary, how to live a long healthy life, that kind of style books. But the islands of the southern end of Japan have one of the highest life expectancy in the world and they also have a very high number of centenarians. They have a low prevalence of disease relative to other areas in Japan as well as to other countries. So this is a bar graph here just very simply showing that the prevalence of different chronic diseases, so coronary heart disease, colon cancer, prostate and breast cancer in Okinawa, the population there versus Japan and in the US. So you can see they have a strikingly lower prevalence of chronic diseases and especially of coronary heart disease. So there are many possible reasons for the differences in longevity and chronic disease, including genetics and overall lifestyle, but the diet has been particularly of interest among people from Okinawa because it's quite different than the rest of Japan and in other places in the world. So there's as a result been quite a considerable amount of study around the Okinawa diet. So unlike the rest of Japan, purple sweet potato is the main carbohydrate, whereas I think when most of us think about Japanese food we might think about kind of the staple food as being a white glutinous rice. So Okinawa diet as a result is very high in carbohydrates, but it's also very high in fiber and very low in processed foods. So if you think about to what the recommended range of carbohydrates and fat are in the US and Canadian diets, it's usually around 45 to 65% for carbohydrates compared to 85% that we see on average people consuming in the traditional Okinawa diet. And fat is usually around 20 to 35% is the recommended range in North American populations. So you can see quite striking differences. However, what is known as the Okinawa diet is really more traditional diet and dietary intake in younger generations is becoming increasingly closer to a Western dietary pattern, which is high in saturated fat and processed foods. And this is occurring in tandem with increased body weight and increased risk of chronic disease. There are several other dietary patterns that may be associated with longevity, which have also been identified through demographic studies. So for example, it's been observed many decades ago that heart disease was lower in countries bordering the Mediterranean seas. The traditional diet there included daily consumption of fruits and vegetables, whole grains and healthy fat, so predominantly olive oil, as well as weekly intake of fish and poultry beans and eggs, very moderate portions of dairy and very limited amounts of bread meat. So the Mediterranean diet is now one of the most widely recommended for overall health promotion and a reduction of the risk of chronic disease and particularly coronary heart disease. So I won't spend a lot of time going over kind of all the different diets that have been linked to longevity, but just to really mention, this is obviously a very active area of research. So there's been some evidence around a Nordic diet or a caloric restriction diet. So typically around 15% fewer calories than they're recommended for a given age and sex, which is also what's seen in the Okinawa diet. So collectively, the evidence does seem to suggest a link between diet and longevity, although there are lots of reining questions about causality. There are also very few studies in the oldest old, so those that are 85 and older and very few studies in centrenarians. And most of these type of studies in the oldest old and centrenarians have been in populations outside of North America who likely have very different diets than the traditional diets that have kind of been studied. So the oldest old, those again are over 85 are some of the fastest growing segment of the population in some parts of the world. However, a few people live to this age without developing chronic disease. So it's important to understand the influence of diet on the achievement of exceptional longevity and the role, if any, on health span. So studies that characterize dietary intake in such populations will add to our knowledge base and also provide some information for our hypothesis testing. So our aim was to assess the dietary intake of a population of men and women who are 85 and older who were free of chronic disease and compare their dietary patterns to adults who are 65 and older using data from the CLSA. So we hypothesized that the oldest old without chronic disease would have dietary patterns that more closely follow guidelines for chronic disease risk reduction such as more frequent consumption of fruits and vegetables, whole grains and lean protein. The healthy aging study is led by Dr. Angela Brooks-Philson out of the BC Genomic Sciences Center at BC Cancer. So this is a study that was designed to study genetic factors that underlie healthy aging and resistance to age related disease. Between 2004 and 2007, they recruited participants who are age 85 and older from the Metro Vancouver area. Two groups were originally recruited. They had the usual ages group who were not selected for health disease status and they had what they call as the super seniors. So that's the term I'll use throughout my presentation. And the super seniors are those who are 85 and older. The eligibility criteria for them were self-reporting never having been diagnosed or taking medications that were prescribed for cancer, cardiovascular or pulmonary disease dementia or diabetes. And participants as part of the healthy aging study completed health and demographic questions and enrollment but they were not asked about dietary consumption. And the study is not actively following participants but they are always looking for enrollments on just put the email address up at the study coordinator there in case you have anybody who might fit that criteria. So one of the challenges when studying the oldest old or centenarians is defining an appropriate comparator group. The vast majority of individuals in the same birth cohort as the oldest old generally do not survive to an advanced age. And those who do may have other limitations that negate participation in research studies. So the usual ages group from the healthy aging had aged by about 10 years by the time we were starting our ancillary study but was still considerably younger than the super seniors. So they started the youngest age group was around 50. So as a result, we decided to apply to use data from the CLSA since it encompasses the same geographic region as super seniors. And it's more likely to represent the aging population than the smaller usual age control group. So this is the same approach that is used by centenarian studies but it does also introduce potential generational and cultural influences which I'll mention in greater depth later in my talk. So for our study, we recontacted to the 177 super seniors who consented to be recontacted. We mailed the packages to the super seniors containing a questionnaire on demographics and dietary intake from the potential pool of 177 participants, two were not interested, four were deceased, some had moved. We also asked about the development and presence of chronic disease since there had been a time lapse between enrollment in the healthy aging study and in our group. So participants who answered yes to any of the queried incident disease questions were subsequently excluded. So that left us with a final sample size of 122 of the super seniors. Participants were asked to complete demographic questionnaires which were drawn from the Canadian Community Health Survey. We also asked them to complete the short dietary questionnaire. So it was the same one as in the CLSA. So we received permission and resources from Dr. Breyna Shantenstein to use the SDQ that was developed for the CLSA. So the SDQ, and I won't go into too great a depth because I think some of you might be familiar with it and Dr. Hurley also described this as well. But briefly, it assesses usual consumption frequencies in the last 12 months of key nutrients and foods that are important for health promotion and chronic disease prevention in younger and older adults. So it has been tested for using community dwelling adults and it's been validated relative to three 24-hour dietary recalls. But notably questions about portion size are not included in the SDQ. So we use data from the CLSA, a subset of 30,000 individuals are also known as the comprehensive cohort. So these are the participants who underwent the face-to-face interview questionnaires, which included dietary assessment. Baseline data was collected over three-year period and completed in 2015, which is slightly earlier than our data, which was collected at the beginning of 2017. So data was obtained from CLSA participants with dietary information and covariates that were similar to those that were collected in our study. We can find our analysis to CLSA participants who are aged 65 and older to provide a comparative group of about 12,626 older adults. And we did not apply any additional exclusion criteria to the CLSA data set. Because we're interested in overall dietary intake versus single nutrients and foods, we used a principal component analysis to reduce the 36-item SDQ variables into a smaller set of variables and to identify dietary patterns. The factor loadings represent the relationship of each food or food group to the underlying factor. Two dietary patterns were identified through the use of scree plots. So we identified a pattern that roughly followed that the Western dietary pattern. So the strongest loadings are shown here. So there were things that processed meat and red meat, french fries, sauces and gravy is fried potatoes as follows high sugar snacks and butter. The other one followed more of a nutrient-rich dietary pattern. So there was higher consumption of fruits and vegetables, cold grains, nuts and seeds, fish, as well as salad dressing. So each participant was assigned a factor score for a given dietary pattern and then they were grouped into quartiles. So participants in quartile four of a given dietary pattern had the greatest tendency to follow that diet. We then used multibariable logistic regression to calculate the odds ratios for being a super senior. So quartile one was the reference for each dietary pattern. Our model one was unadjusted and then the change in coefficient method was used to identify significant confounders. And you can see the ones that were included in our model two here. We did not adjust for age in our model since there's minimal overlap between the super seniors and the CLSA group. So this table shows the demographics of the CLSA and super seniors, not surprisingly super seniors were older. So it had a mean age of 90. Both populations were predominantly white and had a similar income categories. Super seniors were more likely to be living alone. So nearly 64% reported living alone. They also had a lower education were more likely to be never smokers and none were current smokers. Conversely, there were more likely to report habitual alcohol consumption. And another striking difference I did want to highlight was their BMI. So they predominantly had a normal BMI was only about one third reporting being overweight or obese compared to about 70% of CLSA participants. So this table shows the unadjusted and adjusted odds ratios for being a super senior. So the highest quartile of the Western dietary factor was associated with greater odds of being super seniors. And after adjustment in model two, the these associations were strengthened and no observations were observed for kind of the middle quartiles of two and three. Similarly, the highest quartile of the nutrient rich dietary factor was also associated with greater odds of being a super senior. However, those associations were attenuated when we accounted for covariates. And of the covariates, smoking, BMI and alcohol consumption had the largest impact on the effect estimates. So the highest quartile of the Western dietary pattern was associated with the greater odds of being a super senior. So this is really in contrast to our hypothesis where we thought the Western dietary pattern would be more likely to be prevalent in the comparator group. Since this was kind of characterized by less healthy foods, so greater intake of processed meat, red meat, sauces and gravy fried potatoes. Nonetheless, it's important to note that the Western dietary component also contained other factors or other foods, I should say that did not meet our threshold for component loading in the PCA analysis. So things like poultry and eggs that can still contribute to associations even though they're kind of below that statistical threshold. So our apparent finding of a disconnect between the kind of the, where the less healthy Western dietary pattern and the nutrient rich dietary pattern, at least in unadjusted models were both associated with greater odds of being a super senior is actually similar to some studies that we see from centenarian studies where they report that centenarians have a more varied diet, but then they may also be more likely to consume high sugar foods such as cookies and biscuits and may be less likely to follow nutrition guidelines for chronic disease prevention. So it's unclear though, however, whether the tendency of super seniors to follow a Western dietary pattern reflects generational or cohort differences with CLSA participants or whether it actually truly reflects longevity related differences or possibly fatalism. So approaching the end of life and kind of that, I'm gonna eat chocolate if I wanna eat chocolate. I've already made it to 85, 90 years old. So the higher frequency of high fat foods may also reflect ingrained generational dietary behaviors. So notably national dietary guidelines were not significantly modified until the 1980s to emphasize energy balance and moderation. So things like limited fat, sugar, salt and alcohol were really not part of the guidelines until a few decades ago. There's also potential influence of socio-demographic factors. So education was lower in the super seniors and a higher proportion of them lived alone. So both of which are associated with poor dietary intake. And it's also important to recognize some of the limitations of the SDQ which may impair our ability to draw firm conclusions here. So it's not designed to quantify the amount of food consumed. So it might also be prone to of course than hair bias of self-reported dietary intake. And we didn't measure other factors as well that may influence dietary intake in healthy aging. So such as social factors, physical activity as well as genetic variation which would be informative to assess in future studies. So our results and our inferences are thus a little bit limited and it should be taken in light of some of these limitations specifically around kind of the frequency of consumption rather than amount. And this is an important thing to acknowledge because total food intake does decrease with age. And so this might be something that could be addressed in a future study. So finally, thank you to the investigators, the students and study team that made this work possible as well thank you to the investigators in CLSA for allowing us to use the SDQ in this analysis. And I will stop showing there and it looks like there's lots of questions. So that's great to see. Thank you all. Great. Well, thank you to both of you. The challenge now will be getting through the questions. So I'll try to maybe start with two for each of you and then we'll go from there. Just to note, if we don't get to your questions we can try to address them and get back to you after the seminar. So thinking back to Dr. Hurley's presentation the first question is one evaluating grip strength which differs systematically by sex. Why not do separate models for men and women? Yes, thank you for your question. And that is a great point. I mean, to your point, we found that when we included sex as a covariate in the model it was significant with female sex related to lower grip strength. So it is a great point and we didn't run the analysis differently or separate models I should say by sex but it would be interesting to see whether there would be any reflected differences in the outcomes in terms of both nutrition risk and diet. And then also from Jerry Lynn Pryor who has lots of questions today. Was family history of osteoarthritis or joint replacement related to OA or to mobility? Yeah, also very interesting and unfortunately this wasn't one of the variables that was available to look at. So we didn't look at whether family history again was related to each of those measures. And then maybe just one more quick one. I think there was a note in terms of comparing the results of your study to what's been found already. Maybe you can just touch on one or two of the key learnings that were unique to your study or if you didn't end up finding that those new results maybe why you think that was. Yeah, so absolutely the largest contribution for this research is looking at this in an osteoarthritic population. So there have been relationships demonstrated between again aspects of diet as well as nutrition risk among older adults and again relating it to things like physical function and other aspects of health. But this hasn't been evaluated in an osteoarthritic population and particularly given the prevalence of osteoarthritis and the lack of ability to kind of seek out surgical treatment trying to come up with different interventions to potentially improve physical function as well as general health in a way is very important. And so while we've sort of looked at it from an exercise perspective this research contributed great findings regarding potential nutritional strategies that we could use to try and combat osteoarthritic related symptoms again, non-surgical. Great, so maybe we'll go to Dr. Murphy now for a question. What is the comparative rate of diabetes in Okinawa versus USA and Canada? In other words, does the high complex carb diet increase the risk for T2DM or type two diabetes? Yeah, that's a great question. I think intuitively you would look at the composition of the diet and think like the carbohydrates are so high that it may be problematic for things like diabetes, for example, but that's not actually seen, that's not reflected in the prevalence of diabetes in that older population in Okinawa. So they have lower rates of diabetes and I think it's also important to think about kind of although the carbohydrates are quite high, there are lower glycemic index carbohydrate versus like white rice for example, it's also very, very high in fiber. So yeah, that you don't see that relationship there. And another question is, did you look at ethnicity of super seniors or their immigration status? What about which provinces had the most? So the super seniors are only in BC so we didn't look at provincial differences. They're predominantly from the Metro Vancouver area. I mean, certainly we'd love to be able to look at things like ethnicity and immigration status. It is a relatively small population because it's a very unique and interesting group to study but there was only about 8% of people who reported ethnicity that was not white and we didn't ask about immigration status. What about good quality of, and this is also for you as well, good quality of protein in the Western versus healthy diet pattern potentially play a role in the association identified? Yeah, absolutely. I mean, I think that's, this was designed to be kind of a hypothesis generating first look at what might their diets even be because this is a really, as I said, kind of interesting rare group to study. So we were interested to see what they were even eating without even comparing to kind of the CLSA or the usual aging group. That said, I think in hindsight, we chose the SDQ to allow comparison to the CLSA group and also because of the lower burden on this population but it wouldn't allow us to look at more detailed dietary intakes of things like more of the quality of the protein and understand a little bit more of what the actual amounts are as well as since we were limited to frequency but yeah, that's an excellent point. And I'll just maybe ask you this one last question and then we'll go back to Jacqueline. What was the duration of the diet if you collected that information? And since what age was the diet followed? Yeah, that's another important consideration. So this is a dietary intake over the past 12 months. So this might not reflect earlier dietary patterns or things that, so what they consumed kind of earlier in their life that may have contributed to their current kind of disease or lack of disease. So a lot of the dietary patterns do tend to be quite ingrained. I mean, in that they follow them a similar dietary intake kind of throughout time but we don't know that for sure that we're using this single dietary assessment. Okay, so back to you, Dr. Hurley. So Danielle says, thank you Jacqueline for the great presentation and I think lots of our questions started with that. So again, your presentations were both great which is the estimated magnitude of the association between nutrition variables, both diet and nutrition risk and physical activity in general health outcomes. Is it relevant from an epidemiological perspective to justify intervention targeted to diet and nutrition behaviors other than being statistically significant? Yes, thank you. And we didn't look at calculating the correlation between each of our different variables but I think including the effect size can certainly speak to the fact that they would be meaningful besides just being statistically significant. So particularly with nutrition risk and general health and against also sort of seeing these relationships with mobility, again, despite not being statistically significant for one aspect of mobility showing that there is a moderate to high effect size again for self-reported general health for a nutrition risk. So again, certainly suggesting that nutrition risk and again behavior surrounding nutrition would be important to consider for an intervention. And just a note for everyone we have time for a few more questions but if you have to leave a few minutes early if you can just complete the survey on your way out there's the link is posted in the chat box. And now for a question for both of you. Wondering if you use survey weights in your analyses? Maybe. Yeah, would this be? Oh, sorry. No, go ahead. So, yeah, I know we did not use different survey weights in ours, no. No, we didn't in ours either. Sure, three. Okay, I think we have one second. And then in the context of the limitations of cross-sectional epidemiology what role do you think the method of Mendelian randomization can have in identifying causal relationships between diet, nutrition and disease outcome risk? That's a good one. Either of you have a comment on that or have you used that approach before? I haven't used that approach before. So I can't really speak to the effect there. Yeah, so I have used the Mendelian randomization actually not in the context of dietary intake but more in kind of genetic and phenotypic relationships with BMI for example. I mean, I think it's interesting and that's one of the things that we had hoped to look at as well in the healthy Asian study because it was very much developed to be a genetic study. So to understand kind of some of the interactions because I think that, I mean, we were surprised to see our finding that they seem to have less healthy dietary intake. But it is supposed to be an interesting question of what's more important to kind of diet versus genes versus all these other exposures and how do they interact? And it's not something that I think we can look at in that smaller sample size but yeah, I think it's certainly an interesting approach. Okay, great. I think that is all of the questions. Again, if you think of any questions for either of our panelists, you can always email them in and we can get them to them or somehow either to the CLSA or to them directly. But first and foremost, thank you again to both of you for taking the time out of your schedules to do this presentation. We greatly appreciate you participating in the CLSA webinar series and sharing your research. I'd like to remind everyone that the next deadline for data access applications is January 12th of 2022. If you'd like more information, please visit the CLSA website under Data Access to review what data is available, including the COVID-19 questionnaire study data as well as details about the application process. I'd also like to remind everyone to complete their anonymous survey before exiting the Zoom session. We would greatly appreciate that. In terms of our upcoming webinar in December, our final webinar of the year will be entitled Functional Support and Memory, a three-year analysis of the CLSA comprehensive cohort. It will take place on December 17th at noon and presented by Samantha Yu, who's a PhD candidate in epidemiology and public health at University of Ottawa. And you can register for that webinar at the link that is, I believe, on your slide right now. And remember, the CLSA promotes this webinar series using the hashtag CLSA webinar. We invite you to follow us on Twitter at CLSA underscore ELCB. So thank you again to everyone for attending, but also for the panelists today for your presentations.