Digital phenotyping offers a novel and cost-efficient approach for managing depression and anxiety. Previous studies, often limited to small-to-medium or specific populations, may lack generalizability. We conducted a cross-sectional analysis of data from 10,129 participants recruited from a UK-based general population between June 2020 and August 2022. Participants shared wearable (Fitbit) data and self-reported questionnaires on depression, anxiety, and mood via a study app. We examined correlations between mental health scores and wearable-derived features, demographics, health variables, and mood assessments. Unsupervised clustering was used to identify behavioural patterns associated with depression and anxiety. Furthermore, we employed separate XGBoost machine learning models to predict depression and anxiety severity and compared the performance using different subsets of features. We observed significant associations between the severity of depression and anxiety with several factors, including mood, age, gender, BMI, sleep patterns, physical activity, and heart rate. Clustering analysis revealed that participants simultaneously exhibiting lower physical activity levels and higher heart rates reported more severe symptoms. Prediction models incorporating all types of variables achieved the best performance (R2 = 0.41, MAE = 3.42 for depression; R2 = 0.31, MAE = 3.50 for anxiety) compared to those using subsets of variables. Several wearable-derived features were observed to have non-linear relationships with depression and anxiety in the prediction models. Data collection during the COVID-19 pandemic may introduce biases. This study identified several indicators for depression and anxiety and highlighted the potential of digital phenotyping and machine learning technologies for rapid screening of mental disorders in general populations.
Read full abstract