Premium
Causal mediation analysis with longitudinal data: Relationships among sleep quality, diabetes, and subjective memory in older adults
Author(s) -
Park SangMi,
Lee Chang Dae
Publication year - 2021
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.052709
Subject(s) - mediation , cognition , psychology , longitudinal study , sleep (system call) , gerontology , health and retirement study , medicine , clinical psychology , psychiatry , pathology , political science , computer science , law , operating system
Background Low quality sleep or lack of sleep is related to both lower levels of quality of life and negative health problems such as cognitive decline or diabetes in older adults. Although the associations among sleep quality, cognitive function, and diabetes were reported in previous studies, the mechanisms of association and their causal link is unclear. The aim of this study was to explore the mechanism between quality of sleep and cognitive decline by examining the mediation model of quality of sleep, diabetes, and subjective memory. Method The longitudinal panel data from the National Health and Aging Trends Study (NHATS) which was released to the public for research was used for this study. Research hypothesis was that diabetes in the third wave mediates the association between sleep quality in the first wave and subjective memory in the fifth wave among consecutive five years of wave. Presence of diabetes was used as a mediation variable. Sleep quality was measured with three questions about frequency of difficulties in falling asleep or maintaining a good sleep at night within the last month. Subjective memory was measured by one question with a five‐point‐Likert scale. Causal mediation analysis was conducted using CAUSALMED procedure in SAS 9.4. Result Except for the cases which contained missing data, data from 3,238 older adults above the age of 65 was used for final analysis. After controlling for diabetes, education, subjective memory, economic well‐being, depression, medication, age, smoking history in the first wave, the total effect of higher sleep quality on better subjective memory was .011 ( p =.036). However, indirect effect of sleep quality mediated by diabetes was not statistically significant (estimate = ‐.0003, p =.641). The percentage due to interaction effect of sleep quality and diabetes was not statistically significant as well (estimate = ‐31.269, p =.295). Conclusion Although both sleep quality and diabetes were reported to be important predictors of dementia in previous studies, the result did not support the mediation effect of diabetes between sleep quality and subjective cognitive decline. Further study to discover the mediation effect of predictors of dementia is needed for better understanding of the cognitive decline mechanism in older adults.