A remote digital memory composite to detect cognitive impairment in memory clinic samples in unsupervised settings...

Area of Science:

• Digital health
• Neuroscience
• Cognitive assessment

Background:

• Remote monitoring of cognition can aid in early detection of cognitive impairment, especially for Alzheimer's disease patients experiencing memory issues.
• Existing cognitive assessments require in-person visits, posing challenges for widespread case-finding and monitoring.

Purpose of the Study:

• To develop and validate a remote digital memory composite (RDMC) score using an unsupervised cognitive assessment battery.
• To evaluate the construct validity, retest reliability, and diagnostic accuracy of the RDMC score for predicting mild cognitive impairment (MCI).

Main Methods:

• 199 participants (healthy controls, subjective cognitive decline, MCI) completed unsupervised cognitive assessments via a smartphone app.
• A remote digital memory composite (RDMC) score was derived, focusing on episodic memory and long-term recall.
• RDMC score was correlated with the PACC5 score and assessed for diagnostic accuracy in differentiating impairment.

Main Results:

• The RDMC score showed significant correlation with the PACC5 score and good retest reliability.
• High diagnostic accuracy was achieved in discriminating memory impairment (cross-validated AUC = 0.83).
• Sensitivity and specificity for detecting impairment were 0.82 and 0.72, respectively.

Conclusions:

• Unsupervised remote cognitive assessments via a digital platform can effectively discriminate between cognitively impaired and unimpaired individuals.
• Remote mobile device assessments are feasible for complementing traditional neuropsychological evaluations of episodic memory.
• This approach supports efforts in remote case-finding and monitoring for cognitive health in research and clinical settings.