Abstract
Background: Frailty indicators can operate in dynamic amalgamations of disease conditions, clinical symptoms, biomarkers, medical signals, cognitive characteristics, and even health beliefs and practices. This study is the first to evaluate which, among these multiple frailty-related indicators, are important and differential predictors of clinical cohorts that represent progression along an Alzheimer’s disease (AD) spectrum. We applied machine-learning technology to such indicators in order to identify the leading predictors of three AD spectrum cohorts; viz., subjective cognitive impairment (SCI), mild cognitive impairment (MCI), and AD. The common benchmark was a cohort of cognitively unimpaired (CU) older adults.
Methods: The four cohorts were from the cross-sectional Comprehensive Assessment of Neurodegeneration and Dementia dataset. We used random forest analysis (Python 3.7) to simultaneously test the relative importance of 83 multi-modal frailty indicators in discriminating the cohorts. We performed an explainable artificial intelligence method (Tree Shapley Additive exPlanation values) for deep interpretation of prediction effects.
Results: We observed strong concurrent prediction results, with clusters varying across cohorts. The SCI model demonstrated excellent prediction accuracy (AUC = 0.89). Three leading predictors were poorer quality of life ([QoL]; memory), abnormal lymphocyte count, and abnormal neutrophil count. The MCI model demonstrated a similarly high AUC (0.88). Five leading predictors were poorer QoL (memory, leisure), male sex, abnormal lymphocyte count, and poorer self-rated eyesight. The AD model demonstrated outstanding prediction accuracy (AUC = 0.98). Ten leading predictors were poorer QoL (memory), reduced olfaction, male sex, increased dependence in activities of daily living (n = 6), and poorer visual contrast.
Conclusions: Both convergent and cohort-specific frailty factors discriminated the AD spectrum cohorts. Convergence was observed as all cohorts were marked by lower quality of life (memory), supporting recent research and clinical attention to subjective experiences of memory aging and their potentially broad ramifications. Diversity was displayed in that, of the 14 leading predictors extracted across models, 11 were selectively sensitive to one cohort. A morbidity intensity trend was indicated by an increasing number and diversity of predictors corresponding to clinical severity, especially in AD. Knowledge of differential deficit predictors across AD clinical cohorts may promote precision interventions.
Keywords: Alzheimer’s disease; Canadian Consortium on Neurodegeneration in Aging; Comprehensive Assessment of Neurodegeneration and Dementia; Deficit accumulation; Frailty; Machine learning; Mild cognitive impairment; Random forest analysis; Subjective cognitive impairment; Tree Shapley additive exPlanation values.
Plain Language Summary
Frailty is a well-known and common condition affecting many adults as they age from midlife to their later years. It can result from negative changes in many aging systems, including biological, physical, memory, medical, and even health beliefs and practices. Frailty can affect aging people with relatively normal memory and cognition, and it can also affect people who are living with dementia. We used new information from the CCNA database (known as COMPASS-ND) to study frailty in groups of older adults who represented different phases of Alzheimer’s disease. Specifically, the four groups were older adults who were known to be living with (1) no cognitive impairment, (2) subjective cognitive impairment, (3) mild cognitive impairment, or (4) Alzheimer’s disease. We used new mathematical techniques that consider many factors simultaneously. This allowed us to identify which specific frailty aspects were the most important risk factors for developing subjective cognitive impairment, mild cognitive impairment, and Alzheimer’s disease. We describe two key findings. First, our results showed that some aspects of frailty, such a poorer quality of life, increased risk for the three clinical conditions. Second, our results showed that some aspects of frailty were uniquely associated with risk for subjective cognitive impairment (e.g., inflammation), mild cognitive impairment (e.g., poorer self-rated eyesight), and Alzheimer’s disease (e.g., poorer sense of smell). This research helps us understand that frailty has many aspects that can be separated. And these aspects vary across aging people in different phases of aging and Alzheimer’s disease.