Luule Sakkeus , Tallinn University
Katrin Schwanitz, University of Turku
Liili Abuladze, Estonian Institute for Population Studies, Tallinn University
Health status and social networks have rarely been analyzed dynamically. We explore empirically if and how social network changes relate to the disability dynamics across time among middle-aged and older people in 12 European countries. Data came from 21,514 respondents from SHARE (Survey of Health, Ageing and Retirement in Europe). Respondents were aged 55 years and over and were followed-up over about 7 years. Six social network characteristics – reflecting both constant and varying qualities of the network – were used, based on respondents’ own assessment in wave 4 and 6 of the survey. Multi-state modelling was used to investigate links between social network characteristics, health state transitions over time, and death. Individuals with a larger social network size at baseline and those having close emotional ties to other network members had a decreased risk of health decline (hazard ratios 0.96 and 0.95) – controlling for age, gender, education, and country of residence. A social network containing friends at baseline is linked to health recovery (hazard ratio 1.10). Compared to models where each social network measure was entered separately, co-adjusting for all social network measures lead to a change in the statistical significance of the association of having family members as a part of the social network. In all, there is some first evidence that social network characteristics are linked to the disability dynamics of older Europeans. In further analyses we will refine the multistate models, conduct further robustness checks, and investigate cross-national differences in the illness-death process.
Presented in Session 57. Social Networks and Social Support among Older People