Emotional expressivity in early adulthood as a predictor of dementia and Alzheimer's disease in late adulthood
Morrison, Jill M
MetadataShow full item record
Background: As the global population ages, the prevalence of age-related disorders, such as dementia, is increasing. Dementia is a condition of progressive deterioration of cognitive ability that leads to functional deficits. The primary subtype of dementia is Alzheimer’s disease (AD). Currently, there is no cure for AD or other major forms of dementia (e.g., vascular dementia) so prevention is the best approach for reducing the burden of these conditions. Emotions experienced across the lifespan may affect the development of dementia and AD, given their high involvement in cognitive, cardiovascular and psychosocial processes. Emotions may be neuroprotective by promoting the development of cognitive resources that allow resistance to pathologic changes in the brain. Alternatively, emotions may be neuropathogenic by contributing to vascular risk factors and evoking the stress response. The objective of this study is to investigate a potential novel association of emotional expressivity in early adulthood with dementia and AD in late life. Methods: Data from the Nun Study, a longitudinal study of 678 religious sisters who were aged 75+ at baseline in 1991, were used for the investigation. Data include annual cognitive and physical assessments, post-mortem brain autopsies and historical documents obtained from convent archives. Archival autobiographies handwritten in early adulthood (mean age=22) were available for 180 U.S.-born participants. Autobiographies were scored for emotional expressivity, as well as for idea density, a measure of written language skills known to be associated with dementia and AD. Emotional expressivity was classified as high (i.e., top two quartiles) or low (i.e., bottom two quartiles) based on within-convent ranking of number of emotion words. Dementia was diagnosed if individuals displayed an inability to perform activities of daily living, and cognitive impairment on a battery of neuropsychological tests, according to standard criteria. A diagnosis of AD required evidence of dementia and AD neuropathology. Samples were selected for the analysis of dementia (n=149) and AD (n=85) based on the availability of data on dementia, AD neuropathology and all covariates of interest, and on restriction by low education. Positive, negative, and overall emotional expressivity (i.e., the sum of positive and negative emotion words) were investigated in association with both dementia and AD using multivariate logistic regression analysis. Additional analyses were performed to investigate the association of emotional expressivity with dementia. These included dividing the negative emotional expressivity variable into three (as opposed to two) categories, and testing the interaction between positive and negative expressivity in association with dementia. All final models were stratified by idea density and adjusted for age and apolipoprotein E-ε4 (APOE-ε4). Results: The association of emotional expressivity with dementia and AD was modified by idea density. Among individuals with high idea density, those with high emotional expressivity, regardless of valence (i.e., overall, positive and negative), were consistently at an increased risk of dementia and AD compared to those with low emotional expressivity. In particular, overall emotional expressivity was significantly associated with dementia in this subgroup (odds ratio [OR]=2.60, 95% confidence interval [CI]=1.04-7.11). Among individuals with low idea density, those with high overall and negative emotional expressivity were at a decreased risk of dementia and AD compared to individuals with low emotional expressivity. Positive emotional expressivity was associated with an increased risk of dementia and AD. None of the associations of emotional expressivity with dementia or AD were significant in the low idea density subgroup. The associations did not reach statistical significance among individuals with low idea density in the additional analysis of emotional expressivity with dementia. However, among individuals with high idea density, moderate, but not high, negative emotional expressivity was associated with an increased risk of dementia (OR=3.59, 95% CI=1.13-11.89). Furthermore, high negative emotional expressivity was associated with an increased risk of dementia among individuals with low positive emotional expressivity (OR=8.17, 95% CI=1.66-58.96). Discussion: The results support emotional expressivity in early adulthood as a potential predictor of dementia and AD in late adulthood. Idea density, a known risk factor of dementia and AD, modifies the association. High emotional expressivity, regardless of valence, is associated with an increased risk of dementia and AD when cognitive risk is otherwise low (i.e., high idea density), whereas overall and negative expressivity are associated with a decreased risk of dementia and AD when cognitive risk is high (i.e., low idea density). Furthermore, as predicted, the effect of negative emotional expressivity was modified by positive emotional expressivity: negative expressivity was only associated with an increased risk of dementia when positive expressivity was low, suggesting that positive emotions may counteract the adverse effects of negative emotions. Taken all together, the results provide evidence for a potential association of emotional expressivity in early adulthood with dementia and AD in late life. These findings suggest the importance of emotional expressivity as a predictor of long-term health outcomes, including dementia and AD. As such, emotions may serve as a potential target for future dementia and AD prevention strategies.