Latent Cognitive Profiles and Demographic Determinants in Older Adults without Dementia: A National Cross-Sectional...

Traditional cognitive well-being classification in older adults often overlooks domain-specific deficits. We aimed to validate previously identified distinct cognitive profiles among those without dementia and examine demographic predictors to guide tailored interventions for equitable healthy aging.

Cross-sectional study.

Community-dwelling adults aged ≥65 years without dementia from the National Health and Aging Trends Study (NHATS).

We performed a split-sample validation analysis on the identified latent profile analysis of episodic memory, executive function, orientation, psychomotor function, visual attention, and working memory. Multinomial logistic regression examined demographic predictors of each profile with Holm-Bonferroni correction.

After the split-sample validation analysis, we confirmed that 5 latent profiles emerged (n = 2219): profile 1, overall intact (50.5% of all samples); profile 2, isolated moderate orientation impairment (15.6%); profile 3, mild global impairment with preserved orientation (22.0%); profile 4, mild global impairment with significant orientation impairment (5.5%); and profile 5, moderate global impairment (6.2%). Compared with profile 1, profile 2 was exclusively related to aged ≥90 years, with normal income negatively associated. Similarly, profile 3 was linked to ages 85-89 and ≥90 years, as well as Black or Hispanic ethnicity. Profile 4 was positively associated with aged ≥90 years and Black or Hispanic ethnicity, and negatively associated with normal income. Profile 5 was associated with aged ≥90 years and Black or Hispanic ethnicity, whereas normal income and self-rated good or excellent health were negatively associated with profile 5 (all adjusted P < .05).

About half of older adults exhibit domain-specific deficits, suggesting tailored interventions are essential. Profile 5 needs comprehensive strategies, whereas orientation-focused training could benefit profiles 2, 3, and 4. Demographic factors-particularly age, ethnicity, and income-should guide customization to enhance intervention participation and adherence.