Persistent functional impairment as a preclinical signal of Alzheimer's disease: Advancing dementia prognostication through the natural history of function

The field of Alzheimer's disease research has long prioritized cognition as the principal gateway to diagnosis, staging, and risk stratification. While this approach has yielded important insights, it has also delayed early detection by overlooking the behavioral and functional manifestations that often precede measurable cognitive decline.^1–3 In this context, the article by Ghahremani and colleagues, persistent functional impairment as an early indicator of cognitive decline and dementia in cognitively normal older adults, represents a methodologically rigorous and conceptually important advance in dementia research. By demonstrating that persistent functional impairment (FI), rather than transient difficulty, confers more than a twofold increased risk of incident cognitive decline and dementia, this study reframes function as a preclinical signal rather than a late consequence of disease. ⁴

From a social and behavioral public health perspective, the central contribution of this work lies in its attention to the natural history of function. Functional abilities, particularly instrumental activities of daily living (IADLs) are embedded within social context, shaped by routine, environment, cultural expectations, and compensatory support.^2,3 Unlike cognitive test scores obtained in artificial settings, functional behaviors unfold in real life and reflect an individual's capacity to navigate complexity, adapt to demands, and sustain independence. The author's decision to distinguish persistent from transient FI acknowledges this lived reality and aligns function assessment with how neurodegenerative disease progresses.^5,6

The methodological rigor of this study strengthens its impact. Using data from 11,793 cognitively normal older adults in the National Alzheimer's Coordinating Center (NACC), the authors employed exploratory factor analysis to isolate Functional Activities Questionnaire items that more closely reflect functional independence rather than direct cognitive performance. The resulting composite preparing hot drinks, preparing balanced meals, shopping alone, and traveling captures IADLs that are ecologically valid, observable, and meaningful to patients and families.^1,7 Importantly, the authors further minimized cognitive confounding by adjusting for a modified Clinical Dementia Rating (CDR) score that excluded functional domains. This careful separation of function from cognition is rarely achieved in observational dementia research and substantially enhances interpretability.

Equally important is the actualization of persistence. By defining persistent FI as impairment present at more than two-thirds of visits prior to outcome, the authors move beyond static measurement and instead capture trajectory. This approach mirrors recent advances in neuropsychiatric symptom research, particularly the validation of mild behavioral impairment (MBI) as a pre-dementia risk state. ⁸ The parallel is not incidental; instead, it reflects a broader shift toward recognizing that enduring change, not episodic disruption, signals underlying neurodegeneration. In doing so, this study provides compelling evidence that functional persistence belongs alongside cognitive and behavioral persistence in preclinical Alzheimer's disease frameworks.^7,9

The findings also carry significant public health implications. Persistent FI was predictive of dementia even after adjustment for neuropsychiatric symptoms, APOE ε4 status, and subtle cognitive changes. This suggests that functional decline is not merely a downstream manifestation of emerging cognitive impairment but may represent an independent or synergistic pathway of risk.^5,8,9 For population level screening, this is critical. Functional assessment tools are low cost, scalable, and can be administered remotely or by informants, features that biomarkers and advanced neuroimaging lack.^9,10 Incorporating persistent FI into routine primary care or community health assessments could substantially expand early detection, particularly in resource-limited or underserved settings.

Importantly, persistent functional impairment can also be conceptualized as complementary to biomarker-based staging frameworks such as AT(N). The National Institute on Aging-Alzheimer's Association research framework defines Alzheimer's disease biologically by amyloid (A), tau (T), and neurodegeneration (N) biomarker profiles rather than by clinical symptoms alone, reflecting the extended preclinical phase in which pathological changes may precede overt decline. ⁹ Studies show that combining AT(N) biomarkers with cognitive and functional measures predicts future decline more accurately than biomarkers alone, meaning that real-world functional changes add unique and important information. ¹¹ Additionally, staging approaches based on AT(N) principles, such as PET-based Braak frameworks, have been shown to relate to changes in activities of daily living in aging and Alzheimer's disease cohorts, indicating that biological progression is meaningfully linked to functional outcomes. ¹² Tracking functional abilities over time alongside AT(N) markers can therefore improve early detection, make risk assessments more accurate, and better guide both research and clinical care.]

The role of informants, highlighted in the study's findings, further underscores the social dimension of function. The observation that spousal informants were associated with lower detected incidence of decline, while cohabitation increased detection, reflects complex dynamics of normalization, compensation, and observational proximity. These results can remind clinicians and researchers that functional data are relational, filtered through the perceptions and behaviors of those providing information.^2,7 Adjusting for informant characteristics, as done here, should become standard practice in functional and behavioral dementia research and clinical practice. ¹³

Despite the study's strengths, it also exposes ongoing challenges in functional measurement. The Functional Activities Questionnaire, though widely used, was developed for dementia populations and does not fully capture modern IADLs such as technology use, digital navigation, or complex multitasking-domains increasingly central to independent living.^1,7 Additionally, the authors acknowledged that the demographic composition of NACC limits generalizability. Functional expectations, household roles, and interpretations of “difficulty” vary across cultures, socioeconomic strata, and gendered norms. For persistent FI to serve as a truly equitable early marker, culturally sensitive validation across diverse populations is essential.

Nonetheless, the translational value of this work is substantial. Persistent FI offers a pragmatic signal that can inform clinical trial enrichment, allowing for the identification of high-risk individuals before cognitive decline becomes measurable.^5,6 It also provides a framework for targeted interventions such as occupational therapy, environmental modification s, caregiver education, and behavioral supports at a stage when independence may still be preserved. From a prevention standpoint, this aligns seamlessly with public health goals of delaying disability, reducing caregiver burden, and extending quality of life. ¹⁰

In sum, Ghahremani et al. advance this field by demonstrating that how function changes over time, matters more than whether it changes at all. Persistent FI emerges from this work not as background noise, but as a meaningful, measurable, and actionable preclinical marker of Alzheimer's disease risk. ⁴ Integrating this lay understanding of function into research, clinical practice, and public health strategy represents an important step toward earlier, fairer, and more person-centered dementia care.^4,9