Remote symptom monitoring using wearable technology: the relationship between mood, cognition and physical activity in depression

Capturing daily fluctuations in mood and cognition can help to more accurately monitor treatment response - but is this feasible for patients with depression?

In research presented at CNS2020, we examined the feasibility of high-frequency cognitive assessments, and their relationship with daily measures of depressive symptoms and physical activity.

Background

Cognitive symptoms are often under-recognised in depression, but can severely impact patients’ clinical symptoms, quality of life and risk of relapse
Fluctuations of cognitive ability and depressive symptoms are interrelated processes requiring a higher fidelity and frequency of measurement
Here we demonstrate the feasibility of remotely collecting cognitive data in individuals living with depression, as well as the relationship of these high-frequency cognitive assessments with the remote monitoring of depressive symptoms and physical activity

Methods

Using an apple watch, thirty adults with mild-to-moderate depression conducted brief, daily assessments of:

Cognition: N-back brief cognitive assessments (2-Back / ~30sec).
Mood: Self-reported
Activity: Daily total step count extracted

Technological solutions

Results

Validation

Daily cognitive tests and mood questionnaires showed good correspondence with widely used and validated measures
Adherence was excellent with participants completing mood questionnaires on 94% of days and cognitive assessments on 96% of days

Covariation between mood, cognition and activity

We observed covariation of Daily Mood & d-prime (p=0.01) and covariation of Daily Mood & Step Count (p=0.0001)
Lagged analysis showed that an increased step count was associated with better mood on the following day (p=0.05)

Conclusions

These data support the feasibility of deploying remote symptom monitoring techniques via wearable technology in psychiatric populations, such as depression
This work establishes methods for synthesizing high-frequency cognitive data, brief mood and biometric data in order to create sensitive digital profiles of an individual’s clinical symptoms