Accelerating real-world prediction and research in Alzheimer's: The M3AD study

Area of Science:

• Gerontology and Public Health
• Computational Biology and Bioinformatics
• Epidemiology

Background:

• Chronic diseases, including Alzheimer's disease and related dementias (AD/ADRD), often co-occur, influencing each other's progression.
• Understanding these complex, multimorbid trajectories is crucial for effective disease management and prevention.
• Existing research often overlooks the intricate interplay of conditions and patient life complexities.

Purpose of the Study:

• To develop a dynamic platform for analyzing multimorbidity in Alzheimer's disease.
• To leverage electronic health records (EHRs) for a comprehensive understanding of AD/ADRD trajectories.
• To inform strategies for risk prediction, clinical management, and real-world evaluation of AD/ADRD treatments.

Main Methods:

• Harmonization and federation of a three-city EHR metaplatform encompassing nearly 10 million patients.
• Integration of EHR data with census tract information to contextualize patient health trajectories.
• Multidisciplinary approach employing advanced analytical methods for real-world data.

Main Results:

• Demonstration of the feasibility of a large-scale, federated EHR metaplatform for AD/ADRD research.
• Establishment of a framework to analyze complex, comorbid disease trajectories.
• Development of a data infrastructure to contextualize health data within socio-demographic factors.

Conclusions:

• The M3AD Study provides a novel, dynamic platform for understanding Alzheimer's disease within the context of multimorbidity.
• This approach enables more accurate prediction and tailored management strategies for AD/ADRD.
• Integrating EHR data with socio-demographic information offers unique insights into disease progression and patient outcomes.