Project description
Forecasting depression to facilitate prevention
Prevention is the most effective way to address depression, which affects around 40 million people in the European region. The primary cause of suicide deaths, depression limits a person’s ability to cope with daily life and to fully function at work or school. Studies show only 50 % of patients improve under initial treatment. In this context, the ERC-funded WARN-D project will develop a personalised early warning system to help identify people at risk and who should receive assistance. The project will follow 2 000 individuals in their daily lives for over 2 years. Numerous tools will be used to develop the system for forecasting depression before it occurs.
Objective
Depression is common, debilitating, and often chronic. It severely compromises the capacity for independent living, and is the strongest predictor of suicide. Young people are disproportionately affected, and many will spend over 20% of their lives in a state of depression. Further, only 50% of patients improve under initial treatment. Experts agree that prevention is the most effective way to change depressions global disease burden. The biggest barrier to successful personalized prevention is to identify those at risk for depression in the near future. My proposal aims to solve the challenge who should receive prevention, and when, by developing the personalized early warning system WARN-D. To implement personalized detection, I will follow 2,000 individuals over 2 years, and integrate emerging theoretical, measurement, and modelling approaches from different scientific fields so far unconnected. Regarding theory, I conceptualize depression as a complex dynamical system in which causal relations and vicious cycles between problems can move the system from a healthy to a clinical state, consistent with the Network Approach to Psychopathology that I co-developed. Regarding measurement, I will follow participants in their daily lives, and collect temporal dynamics of bio-psycho-social variables like mood, anxiety, stress, impairment, sleep, and activity via smart-phone based ecological momentary assessment (EMA) and smart-watch based digital phenotype data. I will use dynamical network models to study the relations among problems, and use parameters of these models, combined with baseline, EMA, and digital phenotype data, to construct the prediction model WARN-D via state-of-the-art machine learning models. The interdisciplinary project combines numerous modern tools to develop a tailored personalized early warning system that forecasts depression reliably before it occurs, promising to radically transform the science of depression detection.
Keywords
Programme(s)
Topic(s)
Funding Scheme
ERC-STG - Starting GrantHost institution
2311 EZ Leiden
Netherlands