VALIDATION OF DIFFERENT SUCCESSFUL AGEING CONSTRUCTS AND COMPARISON OF THEIR PREDICTORS

Abstract

The lack of consensus on what constitutes successful aging (SA) has triggered over 100 different operational definitions. This heterogeneity hinders the straightforward identification of modifiable predictors of SA. It remains to be clarified whether predictors of SA differ as a function of varying constitutes of SA and whether reserve models from dementia research can be used to predict SA. It was therefore the aim of the current study to replicate and validate existing SA constructs and compare their predictors in a German subsample of a prospective longitudinal multicenter study.

The sample consisted of 1995 individuals (M_AGE = 83.9 years; SD = 3.36; 66% females). Health-related parameters were measured repeatedly every 1.5 years over 12 years. Data assessed at follow-up 3 and 5 were used for current analyses. A confirmatory factor analysis (CFA) was performed to test whether different constitutes of SA, such as physical and cognitive health, functioning, well-being and social integration, form a uniform SA construct or represent separate aspects.

Preliminary analyses revealed that the best model fit could be achieved with a SA model that combines all SA constitutes (SA-ALL). If the SA model lacks the constitute for well-being, it showed a lower correlation with GP-rated health status. If the SA construct lacks constitutes for physical health, prediction of mortality over a three years period was significantly worse. Motivational reserve and age were significant predictors of SA-ALL and a pure physiological SA construct. Final results will be presented at the congress.