Trends, transitions and patterning in social activity over time among aging women and men: A secondary analysis of the Canadian Longitudinal Study on Aging (CLSA).

Social isolation matters for health and longevity, but little research examines transitions into or out of social isolation or whether transitions are gendered or socially patterned. We described gender-specific trends in breadth and lack of social participation over 6 years overall and by age, country of origin, geographic location, education, wealth, and household income. We used three waves of CLSA data to evaluate changes in social isolation (0-1 activities) and broad social participation (5+ activities) in adults aged 45-75 (n = 24,788), by gender and socio-demographics, in linear and multinomial logistic regressions with post-estimated predicted probabilities. The number of social activities decreased over time, with greater declines for women. About half the sample (more men than women) stayed not highly socially active (<5 activities) and almost 1 in 5 became not highly socially active. Most adults (77 %) remained not socially isolated and 14 % became or remained socially isolated. Women were more likely than men to remain not highly socially active and less likely to have multiple social isolation transitions. Broad social participation changed over time for several subgroups of women and men, with gender differences notable for income levels. Social disparities in social isolation transitions differed by gender only for education. Older age and socioeconomically disadvantaged adults had higher probabilities of becoming socially isolated or becoming less socially active. Findings indicated the diversity of social activities declined as Canadians age into later life and transitions in both social isolation and social participation differed between genders, especially for specific vulnerable subpopulations.
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2024 The Author(s). Published by Elsevier B.V. All rights reserved.)