The Longitudinal Relationships Between Social Isolation, Frailty, and Health Outcomes Among Canadian Older Adults

Abstract Social isolation and frailty are global public health issues that may lead to poor health outcomes. We tested the two following hypotheses: 1) changes in social isolation and frailty are associated with adverse health outcomes over two years, 2) the associations between social isolation and health vary across different levels of frailty. We estimated a series of latent growth models to test our hypotheses using data from the FRéLE longitudinal study among 1643 Canadian community-dwelling older adults aged 65 years and over. Missing data were handled by pattern mixture models with the assumption of missing not at random. We measured social isolation through social participation, social networks, and social support from different social ties. We assessed frailty using Fried’s criteria. Our results revealed that higher frailty at baseline was associated with a higher rate of comorbidity, depression, and cognitive decline over two years. Less social participation at baseline was associated with comorbidity, depression, and changes in cognitive decline. Less social support from friends, children, partner, and family at baseline was associated with comorbidity, cognitive decline, and changes in depression. Fewer contacts with grandchildren were related to cognitive decline over time. The associations of receiving less support from partner with depression and participating less in social activities with comorbidity, depression, and cognitive decline were higher among frail or prefrail than robust older adults over time. This longitudinal study suggests that intimate connectedness and social participation may ameliorate health status in frail older populations, highlighting the importance of age-friendly city policies.

five stage approach for scoping reviews. We performed a search of published peer-reviewed articles available in PubMed, CINAHL Complete, and PsycINFO to identify relevant studies. Two reviewers conducted the screening of titles, abstracts, and full-texts. Of the 452 records identified, 24 were eligible for full-text screening and five articles met the final inclusion criteria. The following four themes were identified: (1) prevalence of QoL-related barriers among unpaid caregivers of older adults with VI; (2) adverse events among unpaid caregivers of older adults with VI; (3) interventions for unpaid caregivers of older adults with VI; and (4) potential impacts of intervention on unpaid caregivers of older adults with VI. These findings reveal a lack of interventions for unpaid caregivers of older adults with VI, despite the prevalence of QoL-related barriers and adverse events. Research addressing these issues are urgently needed.

LONELINESS, SLEEP QUALITY, AND COGNITIVE FUNCTION IN COMMUNITY-DWELLING OLDER ADULTS
Kexin Yu, 1 and Bernadette Fausto, 2 1. USC,LA,California,United States,Newark,New Jersey,United States Loneliness is a risk factor for cognitive decline in older adults, however, the underlying mechanisms are less understood. Individuals who experience frequent loneliness tend to have poorer sleep quality. Empirical evidence supports the influence of sleep on cognitive health. This study examined the possible mediating effect of sleep characteristics on the relationship between loneliness and cognition. The study sample included 557 participants from wave 2 of the National Social Life, Health, and Aging Project who had actigraphy sleep measures (mean age = 73.17, 52.6% female). Loneliness was assessed with the 3-item UCLA Loneliness Scale. Cognitive function was measured with the Montreal Cognitive Assessment. Five sleep quality indicators were objectively recorded with wearable devices: assumed sleep time; actigraphy sleep time; time spent awake after sleep onset (WASO); sleep fragmentation; and sleep percentage (actigraphy sleep/(assumed sleep + WASO)). Path analysis model results show that WASO, fragmentation, and sleep percentage mediate the link between loneliness and cognitive function. Loneliness was positively related to WASO, and WASO was negatively associated with cognition. Loneliness correlated with increased sleep fragmentation which was associated with worse cognitive function. Individuals who had more frequent loneliness had a lower sleep percentage, and sleep percentage was positively associated with cognitive function. Nonetheless, the path from loneliness to these three sleep characteristics became insignificant after controlling for depressive symposiums. Depressive symptoms and fragmentation were found to double mediate the association between loneliness and cognitive function. Sleep and depression could be underlying pathways for the association between loneliness and cognition. Social isolation and frailty are global public health issues that may lead to poor health outcomes. We tested the two following hypotheses: 1) changes in social isolation and frailty are associated with adverse health outcomes over two years, 2) the associations between social isolation and health vary across different levels of frailty. We estimated a series of latent growth models to test our hypotheses using data from the FRéLE longitudinal study among 1643 Canadian community-dwelling older adults aged 65 years and over. Missing data were handled by pattern mixture models with the assumption of missing not at random. We measured social isolation through social participation, social networks, and social support from different social ties. We assessed frailty using Fried's criteria. Our results revealed that higher frailty at baseline was associated with a higher rate of comorbidity, depression, and cognitive decline over two years. Less social participation at baseline was associated with comorbidity, depression, and changes in cognitive decline. Less social support from friends, children, partner, and family at baseline was associated with comorbidity, cognitive decline, and changes in depression. Fewer contacts with grandchildren were related to cognitive decline over time. The associations of receiving less support from partner with depression and participating less in social activities with comorbidity, depression, and cognitive decline were higher among frail or prefrail than robust older adults over time. This longitudinal study suggests that intimate connectedness and social participation may ameliorate health status in frail older populations, highlighting the importance of age-friendly city policies.

AGING LGBT ADULTS' ACCESS TO SOCIAL RESOURCES ACCORDING TO LGBT IDENTITY AND SOCIODEMOGRAPHICS
Krystal Kittle, 1 Kathrin Boerner, 2 and Kyungmin Kim, 3

University of Nevada, Las Vegas, Las Vegas, Nevada, United States, 2. University of Massachusetts Boston, Boston, Massachusetts, United States, 3. Seoul National University, Seoul, Seoul-t'ukpyolsi, Republic of Korea
Research suggests that social resources positively influence the health and well-being of lesbian, gay, bisexual, and transgender (LGBT) aging adults, but their access to social resources may vary according to LGBT identity. Using data from Aging with Pride: National Health, Aging, and Sexuality/ Gender Study (N=2,536), multivariate models tested how access to social resources varied by LGBT identity and whether the effect of LGBT identity showed additional variations by sociodemographic characteristics (i.e., age and education) among aging LGBT adults. Lesbian respondents had larger social networks than gay respondents, while gay respondents had smaller networks than transgender respondents. Lesbian respondents had more social support and community belonging than other identity groups. Bisexual male respondents and transgender respondents had less support than gay respondents and bisexual male respondents reported less community belonging than gay respondents. Education and age moderated the association between LGBT identity and social support. Findings highlight the importance of considering social support separately from social network size with the understanding that large social networks do not necessarily provide ample social support and this distinction was particularly relevant for transgender respondents who had larger social networks, but less social support than gay respondents. Results also suggest that feelings of LGBT community belonging vary among LGBT identity groups. Health and human service professionals should not only consider the sexual and gender identity of their aging LGBT clients, but also consider the clients' additional sociodemographic characteristics when assessing their access to social resources.

COMPARISON OF TWO SURVEYS USING THE SEXUALITY ASSESSMENT TOOL (SEXAT) IN FLANDERS
Els Messelis, 1 Michael Bauer, 2 Elisabeth Vander Stichele, 3 and els Elaut, 4 1. LACHESIS,Ghent,Belgium,2. La Trobe University,Melbourn,Victoria,Australia,3. AZ Sint Jan,Brugge,Belgium,4. UGent,Gent,Belgium From 2015 it is mandatory in Flanders, Belgium, to develop a policy to deal with sexual abuse in elderly care. Residential Aged Care Facilities (RACF'S) try to focus on this mandatory, but should also pay attention to implement an overall Sex and Intimacy Policy. This study contains a Comparison of two surveys (Messelis &Bauer, 2020 andVander Stichele, e.a. 2020) in Flanders, Belgium, both using the Sexual Assessment Tool (SeAT, Messelis & Bauer, 2017). Both studies aimed to assess how supportive residential aged care facilities are of residents' sexual expression. In the survey of Messelis & Bauer 750 aged care facilities were contacted in 2017-2018 and 69 (9,2%) completed the SexAT survey after three reminders. Vander Stichele e.a. contacted 100 aged care facilities managers in 2019. Twenty of them (20% response rate) completed the SexAT after three reminders. Findings of the Messelis & Bauer survey indicate that 70% of the facilities rated 'very good' to 'good' (score between 21-59/69), while Vander Stichele e.a. found a prevalence of 76% of this score. Both found no facilities were rated 'excellent' (score greater than 60/69). In the category 'improvement needed' (score less than 20/69), percentages were 30% and 23%; a difference of 7% (CI95% of difference in percentage includes zero, not significant). There is room for improvement in residential aged care facilities for the support of sexual expression of residents. The more recent study confirms results of the previous one, and no significant evolution was observed in two consecutive cross-sectional surveys. Prior studies have reported barriers to meeting the sexual needs of older adults within skilled-nursing facilities, such as a lack of privacy, lack of supportive practices and policies, and judgement or discomfort on the part of the staff (Doll,