Researchers at UCLA Health have discerned four unique routes that contribute to the onset of Alzheimer’s disease through the analysis of electronic health records, providing novel insights into the progressive nature of the disorder rather than attributing it solely to individual risk factors.

The research, published in eBioMedicine, analysed longitudinal health data from over 25,000 patients in the University of California Health Data Warehouse and corroborated findings in the nationally diversified All of Us Research Program. In contrast to earlier studies that concentrated on singular risk variables, the UCLA analysis delineated sequential diagnostic patterns that elucidated the progressive trajectory of disorders leading to Alzheimer’s disease.

“We found that multi-step trajectories can indicate greater risk factors for Alzheimer’s disease than single conditions,” stated the first author, Mingzhou Fu. She is a medical informatics pre-doctoral student at UCLA. “Understanding these pathways could fundamentally change how we approach early detection and prevention.”

The Four principal trajectory clusters

• Mental health trajectory: Psychiatric disorders resulting in cognitive deterioration
• Encephalopathy pathway: Conditions of brain dysfunction that progressively worsen over time.
• Mild cognitive impairment pathway: Gradual progression of cognitive loss.
• Vascular disease pathway: Cardiovascular disorders that elevate dementia risk

Each pathway exhibited unique demographic and clinical traits, indicating that various groups may be susceptible to varied development trajectories.

The research revealed that roughly 26% of diagnostic trajectories exhibited consistent directional ordering. Hypertension frequently precedes depressive episodes, thus elevating the risk of Alzheimer’s disease.

“Recognizing these sequential patterns rather than focusing on diagnoses in isolation may help clinicians improve Alzheimer’s disease diagnosis,” said lead author Dr. Timothy Chang, assistant professor in Neurology at UCLA Health.

When assessed in an independent cohort, these multi-step trajectories predicted the likelihood of Alzheimer’s disease with greater accuracy than individual diagnoses alone. This discovery indicates that healthcare practitioners might utilise trajectory patterns for:

• Improved risk stratification: Early identification of high-risk patients in disease progression
• Focused interventions: Disrupting detrimental sequences before their progression
• Customised prevention: Adapting tactics according to individual route patterns

The validation within the All of Us Research Program—a varied, nationally representative cohort—affirmed that similar trajectory patterns are applicable across all populations and demographics.

Methodology of Research

The researchers examined 5,762 patients who provided 6,794 distinct Alzheimer’s progression trajectories. Employing sophisticated computational techniques such as dynamic time warping, machine learning clustering, and network analysis, researchers delineated the temporal links among diagnoses that precede Alzheimer’s disease.

Original Publication
Authors: Mingzhou Fu, Sriram Sankararaman, Bogdan Pasaniuc, Keith Vossel and Timothy S. Chang.
Journal: EBioMedicine
DOI: 10.1016/j.ebiom.2025.105831
Method of Research: Data/statistical analysis
Subject of Research: People
Article Title: Identifying common disease trajectories of Alzheimer’s disease with electronic health records
Article Publication Date: 30-Jun-2025
COI Statement: The authors have no conflicts of interest to disclose related to the content of this manuscript.