Scalable Biomarkers And Time-Series Microsignatures Of Relapse In Major Depressive Disorder

Major depressive disorder (MDD) is a major global health problem that causes recurring episodes of depression. One of the most burdensome parts of MDD is the fact that episodes often come back ("relapse"), even after they are successfully treated. If we could predict relapse, then we could catch and treat episodes early. In a previous smaller study, we found some highly promising markers of relapse in depression. The first involved how tightly "coupled" someone's depression is with their anxiety. The second is a measure of how easily symptoms of depression influence each other. The last one was a measure of the randomness in activity patterns, which we measured using wearable actigraphs (like a "FitBit"). We now need to do a larger study in order to ensure that our ability to predict relapse in depression is reliable and to determine whether it could work clinically. Our study will follow 252 patients across Canada who were successfully treated for their most recent depressive episode. We will follow them for 18 months and measure the amount of time it takes until depressive symptoms return. They will rate their depression and anxiety symptoms every week and have regular clinical assessments every month. They will wear an accelerometer throughout the entire study. Every two months they will also provide speech samples and measures of brain activity. Using these data, we will build a model to predict the time until relapse for individual patients. An important strength of our model is that it will identify which patients are at the greatest risk of relapse, in addition to providing early warning signs of relapse within the next few days to weeks. Ultimately, our study will help us identify which patients will need more treatment or follow-up after successful treatment of depression. This will help us reduce the burden of depression for patients in the long-term, and may also help us to design new treatments to maintain wellness in people with depression.