Lifelong cognitive reserve, dementia, and mild cognitive impairment among older adults with limited education: a population-based cohort study

Abstract Background Early-life educational attainment represents a major proxy for cognitive reserve (CR) that could benefit late-life cognitive function. We sought to investigate the associations of lifelong CR capacity with dementia and mild cognitive impairment (MCI) among people with limited education, while paying special attention to subtypes of dementia and MCI. Methods This population-based cohort study included 2127 participants (age ≥ 60 years; 59.4% women; 81.5% illiteracy or elementary school) who were free of dementia at baseline (August-December 2014) and undertook follow-up examinations in March-September 2018. Lifelong CR at baseline was estimated by integrating early-life education, adulthood occupation and marital status, and late-life physical activity, social activity, and social support using the structural equation models. Dementia, Alzheimer’s disease (AD), and vascular dementia (VaD) were diagnosed following the international criteria. MCI and amnestic and non-amnestic MCI were defined following the Petersen’s criteria. Data were analyzed using Cox proportional-hazards models. Results During the total of 8330.63 person-years of follow-up, 101 were diagnosed with dementia, including 74 with AD and 26 with VaD. Medium and high (vs. low) tertiles of lifelong CR score were associated with multi-adjusted hazards ratio (95% confidence interval, CI) of 0.63 (0.40–1.00) and 0.27 (0.14–0.54) for dementia and 0.60 (0.35–1.02) and 0.18 (0.07–0.49) for AD. The association between higher CR and reduced AD risk was significant in young-old (60–74 years) but not in old-old people (≥ 75 years) (Pinteraction=0.011). Similarly, medium and high (vs. low) CR levels were associated with multi-adjusted hazard ratio (95% CI) of 0.81 (0.62–1.05) and 0.54 (0.40–0.73) for MCI and 0.73 (0.54–0.98) and 0.50 (0.36–0.70) for amnestic MCI. Lifelong CR was not significantly related to VaD or non-amnestic MCI. There was no statistical interaction of lifelong CR with sex and APOE genotype on incident dementia or MCI. Conclusions High lifelong CR is associated with reduced risks of dementia and MCI, especially AD and amnestic MCI. The association of high CR with reduced AD risk existed only in young-old people. This study highlights the importance of lifelong CR in in maintaining late-life cognitive health even among people with limited education.