Abstract
The heterogeneity in the operationalisation of successful ageing (SA) hinders a straightforward examination of SA associations and correlates, and in turn, the identification of potentially modifiable predictors of SA. It is unclear which SA associations and correlates influence all facets of the SA construct, and whether psychosocial reserve models developed in neuropathological ageing research can also be linked to SA. It was therefore the aim of this study to disentangle the effect of various previously identified SA associations and correlates on (1) a general SA factor, which represents the shared underpinnings of three SA facets, and (2) more confined, specific factors, using bifactor modelling. The associations and correlates of three recently validated SA operationalisations were compared in 2478 participants from the German AgeCoDe study, aged 75 years and above. Based on participants' main occupation, cognitive reserve (CR) and motivational reserve (MR) models were built. Younger age, male gender, more education, higher socio-economic status, being married or widowed, as well as more physical exercise and cognitive activities in old age were found to correlate positively with the general SA factor, indicating a simultaneous effect on all aspects of SA. Smoking and ApoE-ε4 were related only to the physiological facet of SA. CR models were significantly related to the general SA factor. Among all SA associations and correlates, proxy indicators of lifelong cognitive activity and physical exercise showed the strongest effects on SA. Future intervention studies should assess the influence of the preservation of active lifestyle across the life span on SA.