Feasibility of identifying factors related to Alzheimer’s disease and related dementia in real-world data

Objective This study aimed to provide a comprehensive understanding of factors associated with Alzheimer’s disease (AD) and AD-related dementias (AD/ADRD), which could aid in studies to develop new treatments for AD/ADRD patients and identify high-risk populations for prevention. Scope and Method In our study, we summarized the risk factors for AD/ADRD by reviewing existing meta-analyses and review articles on risk and preventive factors for AD/ADRD. From this literature review and the identified AD/ADRD factors, we examined the accessibility of these risk and preventive factors in both structured and unstructured Electronic Health Records (EHRs) data. Results In total, we extracted 401 factors in 10 categories from the identified studies. To share our findings, we created an interactive knowledge graph of these risk factors and the relationships among them to assist in the design of future AD/ADRD studies that aim to use large collections of real-world data (RWD) to generate real-world evidence (RWE). Discussion and Conclusion Most factors, including conditions, medications, biomarkers, and procedures, are accessible from structured EHRs. For those not accessible from structured EHRs, clinical narratives serve as promising sources of information. However, evaluating genomic factors using RWD remains to be a challenge, possibly due to the fact that genetic testing for AD/ADRD is still uncommon and poorly documented in both structured and unstructured EHRs. Considering the continuously and rapidly evolving research on AD/ADRD, automated literature mining via natural language processing (NLP) methods offers a way to automatically update our knowledge graph.