Longitudinal wearable monitoring and polygenic score for major depressive disorder in the All of Us Research Program

Major depressive disorder (MDD) susceptibility reflects both stable inherited liability and dynamic behavioral patterns, yet these dimensions are rarely examined together using long-term data in real-world settings. Here, we integrated genomic, electronic health record (EHR), and longitudinal Fitbit wearable data from 3030 adults of genetically inferred European ancestry in the All of Us Research Program, 284 of whom had an EHR-recorded MDD diagnosis during follow-up after a 180-day baseline period. Time-varying Cox models examined associations of the MDD polygenic score (PGS), monthly wearable-derived physical activity and sleep features, and wearable feature × MDD PGS interactions with EHR-recorded MDD. Higher MDD PGS, lower daily steps, lower light and vigorous physical activity, lower sleep efficiency, and greater sleep duration variability were associated with a higher hazard of EHR-recorded MDD. The associations of sedentary time and sleep duration variability with MDD differed across MDD PGS levels, with stronger associations among participants with higher MDD PGS. Model-estimated hazard ratio curves across MDD PGS levels further showed that the behavioral levels corresponding to the same relative hazard differed across levels of genetic liability, with more favorable behavioral levels observed at higher MDD PGS levels. Sequentially integrating MDD PGS, baseline wearable features, monthly wearable features, and selected wearable feature × MDD PGS interactions increased model discrimination, with the C-index increasing from 0.637 to 0.705. These findings support the complementary value of genetic liability and longitudinal behavioral monitoring for characterization of MDD susceptibility and may inform work on genetically informed digital phenotyping for personalized monitoring and prevention.