Mark Eldaief, Massachusetts General Hospital.

MassAITC Cohort: Year 2 (AD/ADRD)

Traditionally, Alzheimer’s Disease and related dementias (ADRD) patients are evaluated infrequently, imprecisely, and in artificial settings. This creates significant barriers to monitoring disease progression and caregiver burden in ADRD, and to assessing the impact of clinical and investigational interventions. 

Digital assessment measures— including digital phenotyping, artificial intelligence (AI) assessments and machine learning algorithms— can provide more reliable, more naturalistic (i.e., in a patient’s home environment) and more frequent (e.g., continuous) measurements in ADRD patients than those used in clinical settings. These assessments may also monitor safety more accurately, assure medication adherence, and facilitate communication with a patient’s medical team. Moreover, because of the enhanced sensitivity of these measures, they may better predict conversion to dementia in preclinical individuals. However, there are barriers to implementing these digital technologies. These include understanding which technologies can be most readily used in older— and not necessarily tech-savvy— populations, understanding the barriers to the adoption of these technologies in these populations, and confirming that these technologies provide tractable and predictive data regarding cognitive and behavioral disease progression.

To address these issues, this project will create and train a “technology-ready cohort” upon which digital assessments relevant to ADRD patients and their caregivers can be tested. Next, we will show that these assessments better predict key cognitive and behavioral outcome measures in ADRD than standard clinical or research visits. In parallel, we will iteratively evaluate and address barriers to the adoption of these technologies in these populations.