Sex Differences in the Associations Between Social Vulnerability, Frailty, 5-year Survival and Long Term Care Home Entry

Abstract ID
3177
Authors' names
J Mah1,2: J MacDonald1; M Andrew1,2; J Quach2, S Stevens3;J Keefe3; K Rockwood1,2; J Godin1,2
Author's provenances
1. Division of Geriatric Medicine, Dalhousie University, Halifax, Nova Scotia, Canada; 2. Geriatric Medicine Research, Dalhousie University and Nova Scotia Health, Halifax, Nova Scotia, Canada; 3. Department of Family Studies and Gerontology, Mount Saint
Abstract category
Abstract sub-category

Abstract

Introduction: Frailty and social vulnerability use deficit accumulation approaches to understand heterogeneity in older adult health outcomes. We examined sex differences in the effect of frailty and social vulnerability on 5-year mortality and long-term care home (LTCH) entry in Nova Scotia, Canada. Methods: We followed community-dwelling older adults 65 years and over who were assessed for public home care supports from 2005 to 2018 using data from the Resident Assessment Instrument. We conducted sex-stratified and sex-disaggregated Cox proportional hazards analyses, adjusting for age, Cognitive Performance Scale and cohort year of entry. Results: Of 5,520 home care clients, mean age was 80.5 (SD 7.5), frailty Index (FI) was 0.23 (SD 0.10) and Social Vulnerability Index (SVI) was 0.22 (SD 0.69). The cohort was 66.6% female who were significantly less frail, more socially vulnerable and more cognitively intact at baseline. At five years, 49.1% females and 63.0% males had died, and 36.3% females and 29.5% males required admission to LTCH. In sex-stratified models, higher SVI was associated with decreased 5-year mortality and increased LTCH entry; while higher FI was associated with increased 5-year mortality and LTCH entry. In sex-disaggregated analyses, higher SVI remained significantly associated with decreased 5-year mortality for females (aHR 0.92; CI: 0.86-0.99, p=0.02), but not for males (aHR 0.94; 0.86-1.02, p=0.11). There was a weaker association between FI and 5-year LTCH placement for males. Conclusion: Greater frailty was associated with LTCH placement and mortality across sexes, as we hypothesized. However, in sex-disaggregated analyses, the association between frailty and LTCH entry was weaker for males and higher social vulnerability was associated with decreased mortality only in female models. This raises the importance of evaluating these populations separately, as well as the question of how current LTCH placement policies may be inadvertently perpetuating the sex (and gendered) differences of aging.