 Again, called social support availability and executive function in the baseline cohort of the Canadian longitudinal study on aging and let me now please introduce our speaker Emily Retter. Emily is a research associate and future PhD student at the University of Waterloo School of Public Health and Health System. Her research interests include social support, loneliness and social isolation and the impact of these constructs on cognitive function in aging population. So, we welcome Emily today and I will turn it over to her. Thank you Jennifer for the introduction. Hi everyone. Thank you for joining me here today. Like Jennifer said, I will be discussing the association between social support and executive function in the baseline cohort. Canadian longitudinal study on aging. So, this is an interesting time to be just looking into the role of social support and social isolation on health. Chances are you're listening to this presentation from your home or from a work that's a lot quieter than usual and you've probably recently been experiencing some disruption to your daily life and rural routines, which may be impacting how you're communicating and connecting with others. And I think though, while we make these important decisions to stay home and physically distance ourselves from others, it's actually a really good time to consider the potential consequences of social isolation on our health. So, with all this happening, why should we be thinking about cognitive function right now? Why is it important? Cognitive function is a collection of mental processes that allow an individual to complete both basic life sustaining and complex tasks and is therefore an important indicator of successful aging. It's important everyday functioning and adaptation to change. And not only is poor cognitive function a problem in itself can also be an indicator of future problems. So, for example, low cognitive function is associated with greater limitations to both ADLs and IADLs, frailty, which is associated with mortality, risk for institutionalization, as well as future dementia and Alzheimer's disease. Of those mental processes that make up your cognitive function, executive function is a key domain and has been indicated as important to successful aging, given its role in the tasks required for daily independent living. Executive function is thought to regulate or control reactions in a given situation while organizing and coordinating a response. It lets you focus on what you need to do when you need to act in a non-automatic way, such as when you're making a decision or shifting between tasks or planning for the future. And executive function is a particular concern for research into cognitive decline because it involves so many different areas of the brain to complete complex tasks. You can imagine if you have a problem in one area of your brain, it could impact the whole process. In general, most people see a worsening of cognitive function over time, although there's great diversity in the amount of cognitive decline each individual may experience. Many risk factors have been found to increase or decrease both the risk and the timing of declines in cognitive function. While some factors such as age, sex, and genetics are not under individual control, many relevant demographic characteristics and lifestyle exposures are considered modifiable, such as socioeconomic status, which includes education and income, physical health issues, such as chronic conditions like diabetes, as well as various lifestyle factors. However, most of these potentially modifiable factors would have to be altered decades before symptoms develop. So, there's a real push to find factors that can be buffered in mid to late life. Social support may be one potential area of intervention for those who may be at a greater risk due to non-modifiable or early life modifiable factors. So, what is social support? Social support is a very complex topic consisting of many different concepts and definitions. At the macro level, social support can broadly be divided into two categories, structural or objective support and functional or subjective support. Structural support refers to objective measures of support, such as marital status, living arrangement, number of friends, relatives, and neighbors. While these forms of support are easily measured and thus more commonly used in previous epidemiological research, structural support may not fully account for how much support the individual actually perceives themselves as having and whether they view that support as important to them. So, for example, a person could be married, have a big family, friends, neighbors, but not feel supported by any of them or a person could be married, but they're supposed to live in another country. Alternatively, a person with a very small social network may feel that their social needs are met. This objective side of support is addressed with the second category. Functional support is a subjective rating of the level of support that individuals perceive as available to themselves and is based on the perception that when social resources adequately or inadequately fulfill their specific social needs. Measure will be discussing social support availability falls into the second category. While there is some inconsistency in past research, generally high social support is associated with higher cognitive function. Satisfaction with social support is associated with better scores on measures of attention, episodic memory, and a lowered risk of developing dementia, whereas low satisfaction increases risk of cognitive impairment and dementia. The somewhat mixed results of past research may be due to inconsistent definitions of both social support and cognitive function, which you can imagine both are very broad topics and can vary quite widely. And two, as mentioned before, many studies have only been able to include structural or objective measures of support, such as marital status, which may not pick up as directly on the level of support people actually have, and that could have weakened the results. So to add to the current evidence of how social support impacts cognitive function, we investigated the impact of low social support availability on one domain of cognitive function, executive function. And we found that low social support availability was significantly associated with low executive function, even after accounting for a wide variety of sociodemographic health and social variables, and that this relationship varied across subtypes of support. So getting into methods, it will come as no surprise that this work was done using the CLSA, an ongoing prospective cohort study of community-going Canadian adults. Participants in the CLSA are recruited into two cohorts, the Tracking and the Comprehensive Cohorts, and the Comprehensive Cohort, which is what we used for this study. Participants complete their assessment at one of 11 data collection centers across seven provinces. For our study, we excluded those who were missing data on any of the chosen variables, as well as those who did not complete their tests at data collection site. For a total analytic sample of 23,491. And only cross-sectional data was available at the time of this research. So our exposure was social support survey from the Medical Outcome Study, which consists of 19 items rated on a scale from one to five. So for each item, participants were asked how often that kind of support was available to them when they needed it. So for example, how often did you have someone to turn to for suggestions about how to deal with a personal problem? Participants then selected an option from one, none of the time, up to five, all of the time, with a higher score indicating greater levels of support. And social support availability can be broken down into several subtypes of support. First, we have tangible support, which is very physical support, such as, do you have someone to take you to the doctor if you needed it? Do you have someone to help you if you were confined to bed? What about help with chores if you were sick? And I'm sure these questions are on a lot of people's minds right now. But what about affection or affectionate support? Do you have someone who shows you love and affection or gives you hugs? Do you have someone that makes you feel loved and wanted? Or emotional informational support? Do you have someone you can count on to listen to you when you need to talk? Or someone to give you advice in a crisis? Finally, positive social interactions. How often do you have someone to get together with for relaxation when you need it? Or someone to have a good time with? And then the items for each of these subtypes are all added together for overall social support, along with one additional item for a total of 19 items. If we look at the distribution of social support in our sample, you can see that across subtypes, people feel they're generally well supported, which is really great to see. As scores on this measure were not normally distributed, however, each subscale and overall social support availability was categorized dichotomously into low support, yes or no. No consistent cutoffs for low support of social support availability were found in our literature for this measure. So we decided on a cutoff of less than or equal to three out of five, based on the distribution of scores. And this was chosen as an indicator of low social support, given the highly skewed distribution, such that only 6 to 11 percent of participants scored under the cutoff on any of the subtypes. And we chose to use an absolute score, three out of five, rather than a relative one, like for example, taking the bottom 25 percent of the sample, so that we had a consistent comparison across subtypes, as well as for comparison across other studies, both our own and outside of our research group. Executive function was calculated using standardized scores on five tests. So scores on all tests are standardized to performance within each test. So the standardized scores are Z scores, which would mean a zero and a standard deviation of one. And then we added them together using this equation here, for an overall executive function score. Stroop is a reaction-based test, so a higher score is actually worse, which is why that Z score was subtracted in the equation and not added. Based on previous literature on cognitive decline and mild cognitive impairment, low executive function was defined as a scores 1.5 standard deviations or greater below the mean. This cutoff was calculated separately in a cognitively healthy sample of CLSA participants. And then that cutoff was applied to our analytic sample to get our two groups of low and not low executive function, which you can see here. Finally, we included a bunch of covariates. Most of them are based on literature, so for example, age group, education, sex, and health variables are very common in the literature. However, there were some additional variables that had not been looked at, but it was thought feasible that they could impact social support or cognitive function. This included things like pet companionship, which I hadn't seen anywhere else, but was available in the CLSA, so we thought that that could definitely be relevant. So getting to our descriptive results, this is more for interest's sake, but if we look at social support availability by age, we can see that it's fairly consistent across age groups, with the oldest age group seeing slightly greater percentages of those in the lower category. Looking at social support availability and executive function, if we just visually compare the distributions of social support in those with low executive function, which is that bottom call row, compared to those without low executive function in the top, we can start to see some differences already. So immediately to me, I see affectionate support. We can see that a more even distribution in those with low executive function compared to higher levels of social support in those who do not have low executive function in that top group there. Definitely more five out of five scores in that group. And positive social interactions as well pops out to me that a lot more people in that five out of five highest level of social support among those who do not have low executive function. We look at the distribution of dichotomous high and low social support by executive function. So that's that cutoff we discussed earlier. We can see that for all subtypes of support, those with low executive function have a greater proportion of low social support. And low executive function is that top row there. And then the bottom row is those with not low executive function. And you can see across the subtypes. Yes, it's very clear. If we look at covariates and executive function for the sociodemographic variables, we can see sex is fairly evenly distributed with low executive function. As we look at age, there is a disproportionate amount of the oldest age group in the low executive function group. So 45% of those with low executive function are in that oldest age group compared to 13% in those with not low executive function. Looking at sociodemographic variables again, education and income vary unsurprisingly and very consistent with previous research compared to those who do not have low executive function. Those with low executive function have a greater proportion of lower education, 4% versus 17%, and lower income. So 4% versus 13%. If we look at health variables, a greater proportion of people with low executive function report health and having at least one chronic condition. Although as you can see the amount of people reporting at least or at least one chronic condition is high across both groups. For additional social variables, a greater proportion of people with low executive function were widowed. See, that's not a big group, but 21% of those with low executive function were widowed. And a lot of them don't have pet companionship, just kind of. And finally, for loneliness, a greater proportion of people with low executive function reported feeling lonely all the time. And these results are consistent with the idea that a variety of social resources do impact our cognitive function. For multivariable results, we found that low social support availability was significantly associated with low executive function, but that this relationship varied across subtypes of support. So we used weighted logistic regression models with odd ratios and 95% confidence intervals to assess the strengths of the associations. So for overall social support and each subtype, we ran four models. In the first model, we ran just support in the outcomes, so that's our crude model. In model B, we added in sociodemographic covariates as well. In model C, we added the health covariates. And in the final model, D, we added social covariates. And we do that in stages, so we can see how the association between social support and executive function changes, as we add in different types of variables. And that way we can get a better understanding of the relationship between social support availability and executive function, because we can see what variables are impacting it. So if we start with overall social support, we can see that in the crude model, model A, those with low overall social support have 2.3 times greater odds of having low executive function compared to those who did not have low social support. But as we added more covariates, the strength of the association decreased from 2.31 over to 1.21. And you can see it's a fairly steady decrease, as we added in more variables. By the time we add in all of those covariates in the final model, the lower bound of the 95% confidence interval dips below one, and is therefore no longer significant. However, the positive direction indicates that those with low social support have greater odds of low executive function. And you can see it's very close to significant, it just was slightly, slightly below. For tangible support, so tangible support with that very physical, do you have someone to get groceries for you? Do you have someone to help you if you are stuck in bed? For that sort of support, it was not significant after the inclusion of health covariates. So we can see it started significant once we included health covariates. We lost significance. But there is still a positive association. And when we ran the model separately for men and women, the association between tangible support and executive function is significant for women, but not in men in the full adjusted model. Affectionate support, if you have someone who gives you hugs, that sort of loving support remains significant even when we add in loneliness and marital status and pet companionship. Which is interesting. We can see that those with low affectionate social support availability had 24% greater odds of having low executive function after we account for everything that we threw in. And emotional and informational support as well, do you have someone who can give you advice on a crisis? So significant. And as we can see, those with low emotional, informational support had 20% greater odds of having low executive function even after we account for all those covariates. And finally, positive social interactions was also significant. So do you have someone you can have fun with and have relax with, enjoy, and I would aim with? That was significantly associated with 27% greater odds of having low executive function if you had low positive social interactions. In terms of covariates, in the final model, so model D, the one we have all our variables in, being younger, female, having a higher level of education, a higher annual household income, reporting good, very good, or excellent, so free to tell, and having a pet who provides you with companionship was significantly associated with lower odds of low executive function. So those were good things. In contrast, being older, male, having lower education or income, reporting at least one chronic disease, having poor self-rated health, and having no pet for companionship was associated with greater odds of low executive function. In terms of the other social variables, both loneliness and being lowed were associated with higher odds of low executive function, so worse, but the results were not statistically significant in those models. So even after accounting for sociodemographics, age, sex, education, income, province, urban, rural residents, health, chronic conditions, self-rated health, clinical depression, and other social variables like marital status and pet companionship and loneliness, low social support availability was significantly associated with low executive function. But as we saw, there were certain subtypes that were significant when others weren't. So discussing strength, we had a very large sample which is unique in these studies. We had 23,491 participants. We had a very wide age range. A lot of past research has been limited to 65 and older, but we were able to include adults in midlife as well. We were able to consider many covariates in one model, which had not been included simultaneously in previous work. So things like pet companionship and even some health variables like self-rated health are rarely included in large studies, so we were very fortunate to have that in the CLSA. We were also very lucky to have multiple tests of executive function, but we were able to combine to get a good measure of a single domain of cognitive function. A lot of studies aren't able to get that specific, so they'll usually have an overall test of cognitive function or they won't be able to get down to the domain level, and that's one reason we see so much diversity in the research in terms of quality results. We were able to include both objective and subjective measures of health as covariates, which again is quite rare in big studies. And most importantly, we could look at a subjective measure of social support availability, which is often not available. So as I said, a lot of studies will collect information on marital status or living arrangements, but a lot of them don't have that measure, so we again were very fortunate that the CLSA thought to include that in limitations. The heterogeneity of the sample increases the risk of confounding factors that we could not account for in the study. The CLSA excluded people with cognitive impairment at baseline, which is, although a limitation, is also supportive for our research for the association between social support and cognitive function, as we still found significant results even with that bottom level of people removed. So within a fairly cognitively healthy sample, we still found that those with lower work executive function had worse social support availability. In terms of selection biases, all participants in the comprehensive cohort come from within 25 to 50 kilometers of 11 data collection sites. So we're only really getting a snippet of the population and that could hinder our generalizability. And because the CLSA is volunteer based, it can be expected there is some self-selection biases, which is particularly relevant for this study because people with higher levels of executive function and higher levels of social support were probably more likely to sign up for study. Finally, only baseline cross-sectional data was available at the time of the study, so we cannot investigate temporality or causation. However, this is being addressed by our ongoing longitudinal work, and I myself am hoping to continue this as part of my PhD. So implications and future directions. Our results indicate that social support availability, particularly affectionate, emotional, informational, and positive social interaction subtypes may be beneficial to cognitive function in middle-aged and older adults. This research adds to existing evidence that psychological and social factors play their role in later life, although social factors are not often prioritized as an indicator of later health. I would doubt that many doctors ask, you know, do you have support? Do you have friends? Do you have family? Et cetera, et cetera. Our results indicate a potential benefit of late social support interventions for cognitive function decline. So strategies directed toward increasing awareness of and access to available social supports may be one potential buffer against age-related cognitive decline. And current services are often directed at increasing tangible support. So for example, meal delivery programs like Meals on Wheels are very tangible support. Yet this is the only subtype of support that was not significantly associated with low-executive function. So there may be room for additional services aimed at increasing other forms of support in order to prevent cognitive decline. So for example, programs that facilitate communal gatherings and relationship building so that people can have someone to relax to or relax with. Someone to give them advice, those sort of things. And I very much recognize that that is easier said than done at the current time, although I'm thinking there's a lot of people working very hard to figure out how to do that in this day and age. Given that social support remains significant, even after we considered all those different variables, we should be considering the health effects of social isolation during times of social and environmental change that can leave a person vulnerable to isolation and limited social connections. So normally in older adults, this could mean during retirement, you know, you're using a lot of connections. Or when children move away, or when you're downsizing houses and moving to a new neighborhood, or the deaths of friends and spouses these times, or perhaps a social support intervention may help buffer any cognitive hits during that time. But this is also something that we have to now consider while living in prolonged isolation during a pandemic. And I would really love to hear your comments and ideas on that, of course, in the chat. So, acknowledgements. Funding for this project was provided by the Canadian Institutes of Health Research. And I would like to specifically thank my supervisor, Dr. Suzanne Tyaz, as well as the research team that I work with, which is led by Dr. Margueritez, and includes Dr. Colleen Maxwell, Dr. Megan O'Connell, Dr. Jean Law, and Dr. Candice Connart. And finally, a real big thank you to the hosts and coordinators of this presentation, Shirley and Jennifer. And now I can take a couple of questions or comments. Great. Well, thank you very much, Emily. That was really informative. And I must say you have a very soft, lovely voice for doing these sorts of webinars. I can see you doing more of this in the future. So hopefully you come back and tell us about your PhD work at some point. Just a reminder, if people can please post any questions that you may have in the cat box, both I will try to raise them, or Emily can read them as you type them. I don't see any just yet. Usually oftentimes people, it starts off sometimes a bit slow as people are shy. I just, one question just to start it off, and this is for my own sake. It's sort of the classic question about what came first, the chicken or the egg. In your, just sort of in terms of background about the overall issue, does executive function often, what's the theory about executive function leading to these outcomes of social isolation you looked at, or is it the reverse where these social factors are affecting individuals executive function? Can you just touch on that? Yeah. So for sure, there's definitely some potential for reverse causation. I'm sure these two both definitely play off each other. One of the big theories I've seen is a stress theory. So stress is obviously very bad for your cognitive function. I think we've all heard that message many times now. And social support kind of buffers that level of stress over time. So when you're going through a hard time, if you're able to turn to others and you have someone you can relax with, you have someone you can talk to, then you're probably going to feel less stress over time. And then you'll take less cognitive hits. That's the big one. There are a few other theories, but that's the stress of offering theories. Great. Thanks. So question from Lauren Bashar. Thanks for your presentation. I'm curious about whether you looked at care use to potentially mediate the relationship between executive function and social support availability. We did not look at care use, but that's definitely something I think in future work would be worth looking at. Because obviously care use is very much tied into social support availability, especially that tangible support. So thank you for suggesting that. And I don't know that you, at least you didn't go over this in your presentation, but did you look at any differences between provinces? I did look at provinces as a covariate. So there was some significant differences. Certain provinces were, did have significant associations compared to others. I didn't want to get too into it just because I didn't have an explanation for it, but there are some differences between provinces. I used Ontario as a comparison group. And I believe Quebec and BC both had significantly higher levels of executive function. That's significantly low. Well, maybe you can come back at some point and talk to us about those inter-provincial differences as a focus of your presentation. What about religious affiliations? I'm thinking up to the questions that we have, and I don't think the CLSA has specific religious affiliations, but did you look at that at all? I don't believe that data was available, but if it was, it was definitely something that we could consider looking at. I mean, there is lots of research on social support and religion. Church is obviously a major support system for a lot of people. So if it was available, we would definitely consider it and maybe in future work it will become available. Okay. And next question. I'm curious if you could differentiate between social, where the question goes, social networks. Where did it go? Social networks, social support, and social participation. It seems that you considered social networks as the structural aspect of social support. How did you measure loneliness? So if you're following along in the chat box, you might find it easier to read some of the questions as well. So, sorry, the question was about how did we measure loneliness? So that was the second part of the question. The first part is, how did you differentiate between social networks, social support, and social participation? So social networks or that structural side of social support. Let me see if I can find the slide. Social support is the objective measures. So structural objective support is that very structural objective, obviously, measures that are kind of factual. So how many friends do you have? What is your marital status? The functional support is how much support you actually perceive as receiving. Like, you have these certain roles in your life. I need someone to talk to, for example, do I have someone who fulfills that role that fills that function in my life? And that is a subjective response. Social engagement is more about your activity within your environment. So do you belong to a church? Do you, how many days do you go hiking with a group? Those sort of things. So how engaged you are in the community? We didn't include those in this one just because it was too complex and I don't know how many of those we had available. But definitely in future work, that's something you consider. In terms of loneliness, loneliness is measured from the depression scale. I believe it's the CESD scale, which is why I included clinical depression and not depression symptoms as a covariate because I had to steal the loneliness scale out of there. And it's, I believe it's loneliness in the last week. How many times you felt lonely in the last week? Okay. Great. Just getting a mark replied to an earlier comment. We would not check for mediation and cross-sectional studies due to questions about the utility of conducting mediation analyses with cross-sectional data. And also there was a comment about support supplied by church groups. So I think that that related to the religious affiliations question. So now from Paul Mick, is there any evidence that social interventions can lead to changes in health outcomes in older people? I wonder how difficult it is to change trajectories in older people. I also wonder if intervening during critical events, as you mentioned, is important? There's definitely literature showing that increasing social support does help buffer cognitive function, especially in women and health in general. I believe there's, I don't want to, not my specific area, but I believe that there's evidence for cardiovascular disease and depression and diabetes. Social support helps buffer that. Don't quote me on that. I think definitely there's a benefit to offering social support during negative events or events where you might be at risk for having lower levels of support. So for example, retirement would be a good time that there could be some way to offer a buffer there. I know a lot of people lose a lot of their support when they leave work. So finding ways that people can connect to other means would be a really useful thing, I think, to both help people have better social support and to potentially improve their cognitive function over time. Great. So Julian Sayle says, great presentation. Thanks. As executive function decreases, latching on to friends and helpers can become obsessive. Even a hug can become seen as an expression of love. So people start to back off. This of course exacerbates the situation with how executive, with low executive functions, sorry. How can one best deal with this? So how can one, if I'm interpreting this correctly, how does one increase social support while executive function decreases? Is that what I'm... Yeah, I think answer it like that and maybe Julian can clarify if that's not the way the question was intended. Okay. I, just from personal, obviously that wasn't covered in this study specifically, but I think, like, suddenly I would say it would be nice if we all had lots of resources available to us, but maybe that's not always realistic. I'm not exactly sure how people increase their social support while they see declines. I mean, there is an increasing number of social groups. I know a lot of older adult communities and are popping up. A lot more groups and churches are down a little bit, I believe, but clubs are definitely increasing. So maybe spreading out your needs across various people would be good. I know that's difficult, but I think that's probably the best answer that I have. I'm sure there are others who know this a lot better than I do. Great. Well, the good answer. So the next question is, congratulations, Emily. I'm a bit curious about the social support from traditional healers for a loved one with dementia in indigenous communities. How do you consider traditional healer support or care as lower or higher social care for executive functions? Again, very, not an area that I'm well versed in. Definitely something that's interesting. Healers in general are very interesting. I would assume if it's someone that you feel increases your support, someone you can turn into for help in a crisis, then it could definitely increase your emotional or informational support, especially. And we saw that that was significantly associated with cognitive outcomes of higher executive functions. So I could imagine that as long as it's someone that you think supports you and you see that as a support that is available to you when you need it, then I can't imagine that it would hurt. So Mark just points out that the only question in CLSA that relates to religious activity is in the social participation module. And it asks about the frequency of participation in religious activities, which is part of structural social support. And another question from Joe, did you find people belonging to clubs, churches, or senior centers do better than the ones that don't? I did not include those in my study. Definitely we would want to look at I believe there is a social participation module that we can include in future work. Again, I can't imagine that if those are kind of a structural measure. So as long as what you are getting from those clubs does increase your actual support you feel you receive, I would imagine that it would improve. It would be beneficial to you. As long as you're actually feeling that, oh yes, this club does increase my support, like it gives me more resources, I imagine that it would be very helpful. And now a question from Rashmi, thank you for your presentation. Would like to know where do received social support falls into the categorization of social support? Also curious to know if social relationships is a broader categorization that includes social support? So received social support as a concept, it would depend on how it's measured. So if it's a question like feel that I receive the support I need when I or I feel that the support I receive fulfills a certain need, that would be functional support. But if you measured it in an objective way, so I have a caretaker that would be structural support. So it would depend on how it's conceptualized in terms of categorization. Social relationships is obviously this broad broad topic. A lot of these terms are words that we kind of use in everyday language, so they kind of have their own meanings. So when we bring them into the research and we give them an operationalization they sometimes change in meaning. Social relationships I'm guessing would be a broader category than social support because I assume it would include all types of social relationships. Okay, so we have a very practical question now. Thank you for your great presentation. I'm new to CLSA. What resources were useful to navigate the CLSA data to understand it, avoid pitfalls and analysis and interpret it correctly? So maybe you can just give your own insight. And then if a man wants to also follow up with us after, we're happy to point them to some resources or if you want to look around on the CLSA website, you'll also find some resources there. Sure, so one I was very lucky to work with Marca Ramis who's very involved in the CLSA, so I kind of had an insider helping me out. But I definitely found the online data portals were very useful where you could look up a variable and see how it was categorized. There's lots of new documents that tell you how people were recruited and how different variables were calculated and I found those incredibly useful as well. Really once you get into it I think it's quite well organized so I think you'll get a hang of it pretty fast. There's definitely lots of resources and you know the like Mark mentions in the chat the study protocols and questionnaires are posted and you can also always send questions to the info at email if you have additional questions and someone will get back to you quickly about that. And the link to the website is posted as well. There was one final I think a question here just another glowing congratulations to you. Your research is very interesting and practical and very well done very good use of the CLSA data set. So you've had lots of lots of good comments. Everyone is so nice thank you. Great job. So we have probably time for a couple more questions. There's one here from another one here from Julian. Lower executive function tends to to become self-fulfilling downward spiral since people find it harder to provide emotional support and tend to back away. It seems that new types of support need to kick in at this stage but it seems there is less support at this stage. Are there any studies about new approaches for people with significantly declining executive function? I am sure they they're definitely is. We were more for this research we were more interested in just the association and really kind of getting an idea of what types of support was associated with cognitive function. But definitely we would probably in the future look at what sort of interventions pick up on these different types of social support and see you know what are useful at what stage. But yes there definitely is those who have cognitive issues or see cognitive declines could definitely experience a closing in of social supports especially when they need it the most. So there is a bit of a feedback loop there as well. I don't specifically know any studies about new approaches for people with significantly declining executive function but I will definitely consider looking into that in the future. So thank you for bringing that up. Okay and next question is how long it does it take to complete the study? I presume referring to your study despite the large population sample. So you've been working on this for a couple years at least. Yeah mine took about two and a half years but the CLSA was very new at that time and we were just figuring things out. A lot of things have been determined now and so for example the executive function cut off that's a lot more commonly well known versus when I was doing this a few years ago that data had just come out so there wasn't any normative data there wasn't a lot of information on that but there's a lot more available now so it would probably go faster. So I guess to wrap up do you what aspects of your study how do you plan on using this research or do you to continue your PhD studies or what will what will that look like? So I found that when I did this I was kind of expecting I think this is how research goes but I was kind of expecting to get a clear answer and that would solve everything and I would know exactly how social support related to cognitive function but what instead happened was I found that I had so many more questions get generated. So if you notice when I mentioned tangible support I talked about when we stratified by sex there was a significant association for women and not men. I ran separate analyses for all the subtypes but because there was so many interactions and there was so many variables that were coming up as important in those associations it was so complex and it was so there were so many different things happening in these that I'm very excited to kind of figure out how to map things out and look at okay these variables are feeding off of each other to find out this is how the various types of social support relate to cognitive function because it seems to be a very complex web of different variables together. So for example marital status and emotional support seem to be very intertwined and pet companionship and affection support seem to be very they're different definitely different variables but it seems to needed to stratify the analysis there. So there's so many interesting things that came up that not I had not come across before so I'm excited to work on it. Yes that's that's the way of research just uh just to note as before participants start to leave I know some people have started to leave to complete the poll that should have popped up on your screen so that we can gather information about future future webinars. I just see one last question from Kathy Fuller and that was about English versus French data and if you looked at language English versus French data also ESL or FSL could it be that executive function undermines quality of communication and that communication function is key to social function? So cognitive function is calculated separately for English speakers and French speakers they have different norms. We did not account for English as a second language although I do know there's research in my other lab going on about language and cognitive function so there is definitely a relationship there. We did not account for it study though but we did separately calculate executive function based on language so that should hopefully be of some hope to you. Okay well great I think the time has come to end the presentation. We really appreciate appreciate your participation in the CLSA Emily into in the webinar series. I'd like to remind everyone that CLSA data access request applications are ongoing. The next deadline for applications is June 17th of this year. Please visit the CLSA website under data access to review available data. Further information and details about the application process. Again I'd also like to remind everyone to complete their survey under the polling option. If you don't see it beside the chat button please click the drop down arrow and our next webinar which will be in June sorry Wednesday May 27th will be Alessandra Andriaki a recent Master of Public Health graduate from McMaster. She'll present on comparing measures of obesity in relation to healthcare use in adults from the CLSA and registration will open for that next week. And to remember the CLSA promotes the webinar series using the hashtag CLSA webinar and we invite you to follow us on Twitter at CLSA underscore ELCV. So again thanks everyone for attending and to Emily and we'll see you next month for Alessandra's webinar.