OBJECTIVE: Longitudinal studies investigating the link between social support and functional decline are limited among the oldest old. Thus, the aim of this study was to examine whether changes in social support are associated with functional decline among the oldest old longitudinally using panel regression models.
METHODS: Longitudinal data from 3 waves (waves 7, 8, and 9) of a multicenter prospective cohort study covering primary care patients aged ≥85 years were used. In the analytical sample, n equaled 624 individuals. The validated Lawton and Brody Instrumental Activities of Daily Living (IADL) scale and the well-established Barthel Index (ADL) were used to quantify functional status. The psychometrically sound Lubben Social Network Scale was used to measure social support. Several potential confounders such as age, marital status, cognitive decline, or depressive symptoms were included in the fixed effects (FE) regression models.
RESULTS: Linear FE regressions showed that a decrease in social support is associated with functional decline (IADL: β = 0.03, p < 0.05; ADL: β = 0.27, p < 0.05) in men but not in women. With IADL as outcome measure, the interaction term (sex × social support) achieved statistical significance (p < 0.01). With regard to covariates, functional decline (IADL and ADL) was consistently associated with increasing age, an increase in the number of chronic conditions (except for women [ADL]), and cognitive decline (except for men [ADL]). Furthermore, functional decline (ADL) was associated with an increase in depressive symptoms.
DISCUSSION: Our findings highlight the meaning of social support for functional status among the oldest old. Finding ways to sustain social support in highest age may be a promising approach in order to postpone functional decline.