There is a lack of studies investigating the link between time-varying factors associated with changes in frailty scores in very old age longitudinally. This is important because the level of frailty is associated with subsequent morbidity and mortality.
To examine time-dependent predictors of frailty among the oldest old using a longitudinal approach.
Longitudinal data were drawn from the multicentre prospective cohort study “Study on Needs, health service use, costs and health-related quality of life in a large sample of oldest-old primary care patients (85+)” (AgeQualiDe), covering primary care patients aged 85 years and over. Three waves were used (from follow-up, FU, wave 7 to FU wave 9 [with 10 months between each wave]; 1,301 observations in the analytical sample). Frailty was assessed using the Canadian Study of Health and Aging (CSHA) Clinical Frailty Scale (CFS). As explanatory variables, we included sociodemographic factors (marital status and age), social isolation as well as health-related variables (depression, dementia, and chronic diseases) in a regression analysis.
In total, 18.9% of the individuals were mildly frail, 12.4% of the individuals were moderately frail, and 0.4% of the individuals were severely frail at FU wave 7. Fixed effects regressions revealed that increases in frailty were associated with increases in age (β = 0.23, p < 0.001), and dementia (β = 0.84, p < 0.01), as well as increases in chronic conditions (β = 0.03, p = 0.058).
The study findings particularly emphasize the importance of changes in age, probably chronic conditions as well as dementia for frailty. Future research is required to elucidate the underlying mechanisms. Furthermore, future longitudinal studies based on panel regression models are required to confirm our findings.

© 2020 S. Karger AG, Basel.