PDF, English - main document Download (1MB) | Terms of use

Abstract

In the face of population aging, the high relevance of identifying favorable conditions, individual resources, and malleable factors to promote cognitive health in old age is unwavering (Deary et al., 2009). Longitudinal research in the last decade suggests that individual beliefs towards age and aging possess remarkable developmental relevance to shaping the actual aging process. As such, evidence underlines the importance of views on aging for health and longevity in old age (Westerhof et al., 2014). Building on the emerging longitudinal work that also suggests associations with memory (Levy, Zonderman, Slade, & Ferrucci, 2012; Stephan, Sutin, Caudroit, & Terracciano, 2016) and pathological brain changes (Levy et al., 2016), the aim of the present dissertation is to explicitly combine the two research traditions of views on aging and cognitive aging. Main reasoning is that negative views on aging constitute a substantial risk factor for cognitive aging that has been rather neglected by the classical literature on age-related cognitive decline. In three individual papers, this dissertation addresses four major questions. First, do people with more negative views on aging show accelerated cognitive decline over time (Paper 1, 3) and have higher odds of developing dementia in old age (Paper 2)? Second, addressing issues of bidirectionality, do cognitive abilities and respective loss experiences—objective change as well as subjective complaints—also contribute to a more negative evaluation of the aging process (Paper 3)? Third, which long-term pathways mediate the relationship between views on aging and cognitive development (Paper 2)? And, fourth, do effects differ depending on individual factors like gender or age group (Paper 1, 3)?

Data came from the population-based Interdisciplinary Longitudinal Study of Adult Development (ILSE; Sattler et al., 2017). ILSE is an ongoing German cohort study starting in 1993/94 with four completed measurement occasions. Two birth cohorts, a midlife group (1950-52; n = 502; Mage = 43.7 at baseline) and an old age group (1930-32; n = 500; Mage = 62.5 at baseline) were reassessed after 4 years (1997/98), 12 years (2005/06), and 20 years (2014/16) with longitudinal response rates of 78.7% and 56.9 % after 12 and 20 years, respectively. Measurements in the present thesis include the widely used Attitude Toward Own Aging scale (ATOA) to assess individual views on aging (Lawton, 1975), well-established cognitive indicators of fluid and crystallized abilities (e.g., WAIS-R; Tewes, 1991), an expert-based clinical diagnosis of participants’ cognitive status, and a range of control variables (e.g., sociodemographics, genetic and health variables, control beliefs, leisure activities). Drawing on the older birth cohorts (1930-32; n = 500) and 12 years of observation in Paper 1, overall and gender-specific latent change score models were applied to investigate the impact of ATOA on rate of decline in fluid and crystallized abilities and whether this impact differed for men and women. In a cognitively healthy subsample of the old age cohort (n = 260), Paper 2 examined whether more negative baseline ATOA increased the risk of developing mild cognitive impairment (MCI) or Alzheimer disease (AD) within 12 years by means of logistic regression. Moreover, leisure activities and control beliefs were examined as possible mediators of the association. Making use of both birth cohorts and 20 years of observation, Paper 3 targeted issues of bidirectionality between ATOA and cognitive development. A multigroup latent growth curve model examined longitudinal associations of ATOA, performance-based cognitive measures, and subjective cognitive complaints contrasting mid- and later life.

First, as expected, negative ATOA was a risk factor for accelerated cognitive decline, predicting change in fluid abilities—but not in crystallized performance—over 12 years (Paper 1) and over 20 years (Paper 3). Moreover, negative baseline ATOA was associated with a 31% higher risk of developing dementia within a 12-year follow-up (Paper 2). These effects emerged after controlling for sociodemographic, various health, as well as genetically relevant indicators and thus seemed to be robust over up to two decades of observation. Second, there was no long-term association between (change in) cognitive abilities and subsequent change in ATOA. Paper 3 revealed instead that cognitive complaints rather than objective cognitive decline were an important correlate and precursor of changes in ATOA. Higher cognitive complaints were substantially related to worsening in ATOA, indicating that the subjective awareness of cognitive loss matters more for attitude formation than objectively measurable decline. Third, testing empirical pathways that were assumed to operate between ATOA and cognitive impairment, evidence for a mediating role of leisure-activity level and control beliefs was scarce (Paper 2). Finally, results illustrated the important consideration of the individual factors gender and age group. Gender-specific analyses showed a stronger association between ATOA and decline in fluid abilities for men, even after controlling for health and education (Paper 1). Moreover, findings indicated that the predictive effect from ATOA on cognitive decline gained developmental relevance with increasing age. That is, ATOA predicted cognitive change over 20 years in old age but not in midlife (Paper 3).

Overall, the results of this dissertation suggest that negative views on aging accelerate cognitive decline and pose a threat for dementia in later life. Taken together with other recent studies, there is robust evidence to conclude that views on aging deserve to be considered as a meaningful risk factor of cognitive decline alongside other established risk factors. This underlines the need of exploiting the emerging potentials and improving views on aging in individuals, communities, and society via targeted interventions and public health practices as a rewarding avenue to promote cognitive health in old age. One important direction for future research therefore is to evaluate whether and how findings can be transferred into efficient intervention strategies. Such interventions on malleable psychological risk factors offer promising low-cost options. Dissertation findings further suggest that interventions may also explicitly treat subjective cognitive complaints. Focusing on individuals already in midlife may prevent them from developing even more complaints and even more negative age views as they age. Another major task for future research is to empirically identify underlying mechanisms, for example, by using measurement burst designs on daily activities and health behaviors to track back what people with more positive views on aging actually do and experience throughout their days. Understanding how negative views on aging accelerate cognitive decline is pivotal to validate and anchor the predictor effect as well as to inform prevention efforts.