A coordinated analysis of the associations among personality traits, cognitive decline, and dementia in older adulthood

Highlights

• Low neuroticism and high openness were associated with high cognitive function.

• Openness was associated with steeper cognitive decline after dementia diagnosis.

• Models were coordinated and harmonized across 4 independent longitudinal studies.

Abstract

There are individual differences in the rates of cognitive decline across later adulthood. Personality traits are among the factors that may account for these differences. The current project investigated whether personality traits were associated with trajectories of cognitive decline, and whether the associations were different before and after dementia diagnosis. The data was analyzed using linear mixed effects regression. Across study aims is a focus on replicability and generalizability. Each question was addressed in four independent longitudinal studies (EAS, MAP, ROS, SATSA), then meta-analyzed, providing estimates of replicability. Results indicated that low neuroticism and high openness were associated with total cognitive function. We detected evidence for cognitive decline in all four samples, and openness was associated with decline post dementia diagnosis.

Introduction

There are substantial individual differences in rates of cognitive decline in older adults. Some persons are able to maintain healthy levels of cognitive ability before experiencing substantial functional loss (Stern, 2012), while others decline more quickly. There is also variation in the timing of decline trajectories relative to dementia onset. The theory of compression of morbidity predicts later onset of dementia along with steeper decline after such onset, thus compressing the amount of time living with severe cognitive impairment (Fries, 2005). There is some empirical evidence that some individuals have slower decline before the threshold of dementia is met, and then faster decline after meeting that threshold (Contador et al., 2017, Yu et al., 2012). This finding is consistent with the compression of morbidity hypothesis, in which the onset of major decline may be compressed into a shorter period before the time of death. Thus, identifying variables related to individual differences in rate of change, as well as timing of cognitive decline is a high priority. Personality traits may be related to both maintained cognitive function and compression of morbidity, given their association with dementia occurrence in later life (Terracciano et al., 2014), as well as other disease and mortality outcomes (Graham et al., 2017, Mroczek and Spiro, 2007). The present study examined associations between personality traits and cognitive decline, and tested whether personality traits are also related to faster decline after dementia diagnosis.

Cognitive decline varies greatly among individuals in old age (Bäckman et al., 2005, Lipnicki et al., 2019, Mella et al., 2018, Zammit et al., 2018). It is well established that most cognitive abilities decrease among healthy older adults, although timing, pace, and direction of change varies (Mella et al., 2018, Salthouse, 2009, Soubelet and Salthouse, 2011, Wilson et al., 2002). However, heterogeneity in cognitive trajectories suggests that some individuals still remain relatively stable or even continue to increase in their abilities throughout midlife (Aartsen et al., 2002, Rönnlund et al., 2005, Schaie, 1989). Accounting for this variability has been a key focus of much of the cognitive aging field over the last several decades.

A high priority for current research in this area is the identification of factors that may be related to, or account for, individual variation in cognitive decline. A recent meta-analysis of previously published work showed support for the associations among education, occupation, and maintained cognitive function (Opdebeeck, Martyr, & Clare, 2016). This active model provides a good framework to help explain why individuals with higher education or occupational achievement are able to maintain better cognitive function before showing clinical signs of decline. Recent work has found that educational attainment and cognitive participation are associated with better cognitive function prior to the onset of decline (Borland et al., 2020, Wilson et al., 2019), and are also associated with faster cognitive decline after a change point (Yu et al., 2012). Some individuals engage in healthful activities throughout their lives and show better functioning, more efficient processing, and slower decline as they age, perhaps a demonstrative of cognitive reserve. These individuals may be more resilient to the brain pathologies that tend to develop with age, may experience slower progression to dementia, but also faster rates of decline after diagnosis, a concept known as the compression of morbidity.

The compression of morbidity is a concept that describes the process by which some individuals maintain functionality throughout old age, and experience steep declines in health and functioning only as they approach end of life. Some individuals retain their cognitive health for a longer period of time, and decline faster after disease onset, thus living longer in a healthy state, and spending a shorter time in a diseased state (Fries, 1980, Fries, 2005). This fits the model of compression of morbidity for cognitive decline and onset of dementia. For some, cognitive decline may be relatively slow and the onset of dementia is delayed, but the rate of decline accelerates after a diagnosis of dementia (Contador et al., 2017, Yu et al., 2012).

The causes of functional loss and death among today’s aging population are primarily in the category of chronic conditions (as opposed to infectious diseases as was the case in prior eras, until recently) (Fries, 2005). As such, the aging population has much more agency in maintaining their quality of life (via preventing chronic conditions) than ever before. Maintaining a healthy lifestyle can help individuals achieve healthy aging, and lengthen the amount of time before the onset of conditions and reduced functioning. This compression is considered preferable, because it allows individuals to have high quality of life and minimizes the amount of time they spend in an impaired state. For gerontologists, understanding how to achieve this compression of morbidity is a key goal for the increasingly large aging population, as it necessarily reduces the burden on the healthcare industry and caregivers, as well as reducing negative stigma about the experience of aging.

There is some evidence for psychosocial factors related to slower decline before the threshold for dementia is met, and then faster decline after. Personality traits may be one of the factors related to both overall cognitive aging and the compression of morbidity. There is a breadth of evidence linking personality traits to cognitive ability and decline across the adult lifespan. Many of these studies have found that individuals higher in openness and conscientiousness, and lower in neuroticism, typically have better cognitive performance (across multiple domains), and less decline over time, see (Curtis et al., 2015, Luchetti et al., 2016) for a review, with somewhat more mixed evidence regarding extraversion and agreeableness. In addition to their associations with cognitive function, personality traits are perhaps most well known in their associations with other aspects of behavior (Mroczek et al., 2009, Turiano et al., 2012), disease (Weston and Jackson, 2015, Weston et al., 2014), and mortality (Graham et al., 2017) at all stages of the adult lifespan. Personality traits are associated with factors that lead to disease or early mortality, most notably health behaviors (Turiano, Chapman, Agrigoroaei, Infurna, & Lachman, 2014) and physician adherence (Hill & Roberts, 2011).

Factors that are associated with less decline include education, occupation, and activity engagement (physical, social, intellectual), all of which are related to personality (Scarmeas & Stern, 2004). Some studies have found that individuals with higher education tend to decline more quickly once dementia is diagnosed (Contador et al., 2017, Hall et al., 2007, Hall et al., 2009, Scarmeas et al., 2006). Given that neuroticism, conscientiousness, and openness also have been linked to health behaviors and education (Chapman et al., 2007, Graham et al., 2020, Hampson et al., 2007), this suggests that personality traits may be a predictor of cognitive decline (or maintenance) and compression of morbidity. Thus, personality traits may be associated with cognitive changes both before and after a dementia diagnosis takes place. Personality traits are associated with patterns of decline (Curtis et al., 2015, Luchetti et al., 2016). Those with higher levels of particular personality traits (e.g., openness, conscientiousness, and lower neuroticism) may have slower rates of cognitive decline, meaning less impairment and better functioning for a longer length of time, but have faster rates of decline after dementia onset. Although this hypothesis has not been tested, indirect evidence from recent studies of older adults have indicated that conscientiousness predicts terminal cognitive decline (Wilson et al., 2015b), and both lower conscientiousness and higher neuroticism are associated with risk of developing Alzheimer’s disease (Duberstein et al., 2011, Terracciano et al., 2014). Neuroticism is associated with higher levels of perceived stress (Jiang et al., 2017) and perceived stress is associated with the onset of amnestic mild cognitive impairment and cognitive decline (Katz et al., 2016, Jiang et al., 2017).

Compressing morbidity into a relatively small number of years in an aging individuals’ life is a goal that envisions optimal health for as long in the life cycle as possible (Fries, Bruce, & Chakravarty, 2011). To achieve this, one strategy is to focus on modifiable lifestyle factors that can reduce or delay the onset of age-related conditions, including cognitive decline and dementia. As discussed previously, personality traits are related to many of these health behaviors, as well as directly related to many age-related health outcomes. As such, personality traits may be directly associated with the compression of morbidity as it applies to cognitive decline. The current study seeks to test this.

The current study expanded upon prior work by examining associations between personality traits and cognitive change in persons both with and without diagnosed dementia. This work was completed in a multi-study coordinated analysis format to assess the replicability and generalizability of the proposed models (Graham et al., in press, Weston et al., 2019). The goal of the current project was to better understand how personality traits were related to the process of cognitive change both before and after a diagnosis of dementia has been received. First, we examined whether personality traits were related to cognitive decline. Specifically, we predicted that lower neuroticism would be associated with less decline in cognitive function. We predicted that higher openness would be associated with less cognitive decline. Lastly, we predicted that higher conscientiousness would be associated with less cognitive decline. Second, we examined whether these traits are related to a deviation in the overall rate of cognitive decline after the development of dementia in persons who were diagnosed with dementia over follow-up. Our hypotheses for these questions reflect that of compression of morbidity. We predicted that individuals lower in neuroticism would experience slower rates of decline overall, and steeper declines after a dementia diagnosis. Similarly, we predicted that individuals higher in openness would be associated with slower rates of decline overall, and steeper declines after a dementia diagnosis. Lastly, we predicted that higher conscientiousness would be associated with slower rates of decline overall, and steeper declines after a dementia diagnosis. The literature is somewhat mixed regarding extraversion and agreeableness, so we did not have a clear hypothesis for these traits. As such, analyses for extraversion and agreeableness were exploratory.