Identifying Cognitive Profiles and Determinants of Cognitive Superaging Among Older Adults

Abstract

Cognitive aging among older adults is heterogeneous and does not follow a single trajectory. This study used a person-centered approach to identify cognitive profiles, specifically cognitive superagers, among United States adults aged 75 and older and to quantify transitions between cognitive profiles across three waves of the Health and Retirement Study in 2012, 2016, and 2020. Five indicators of memory, attention, language fluency, and executive function informed latent profile analyses that yielded four interpretable profiles. These profiles included Cognitive superagers, a Memory strength profile, an Executive Function/Attention strength profile, and a Low cognitive performance profile. Latent transition analysis with measurement invariance showed that the extreme profiles were most stable. Downward transitions were more common than upward transitions, and improvement from the lowest cognitive performance profile was uncommon. Three-step models evaluated profile membership and maintenance by health-related covariates. Older age was strongly associated with lower cognitive profile membership and cognitive decline, whereas higher education and greater physical activity were associated with cognitive superager membership and maintenance of cognitive functioning. Multimorbidity, depressive symptoms, and sleep disturbance were associated with lower profile assignment and increased risk of decline. Sensitivity analyses that addressed survivorship and attrition, including a pattern mixture specification and a stayers-only specification, elucidated the sensitivity of the profile structure on attrition, at baseline, but robustness of the direction of covariate associations. An exploratory analysis in the venous blood study subsample 2016 compared DNA methylation biomarkers across profiles. Cognitive superagers showed modestly high favorable epigenetic aging signals, and the pattern differed across the other cognitive profiles. Findings support a dynamic view of late-life cognition and inform targeted prevention strategies to ensure maintenance of cognitive reserve.