Early Signs: digital phenotyping to identify digital biomarkers for predicting burnout and cognitive functioning in ED clinicians
Brief description of study
Our overall goal is to enhance the ability to assess and understand the development of burnout amongst ED staff using objective, ecologically valid, economical, and accurate digital biomarkers of burnout as a potential alternative to self-report measures for burnout. We are asking ED staff & clinicians who works full-time in the emergency departments to participate. The purpose of this study is to test the feasibility of collecting and analyzing video-recorded semi-structured interviews about experiences at work of Emergency Department (ED) staff to identify digital biomarkers of burnout using Machine Learning methods. As part of the study, we will video-record 15-minute interviews about experiences at work, collect hair samples, blood samples and conduct a neurocognitive test battery (CANTAB®). Afterward, we will use a computer algorithm to test if we can predict work-related well-being (i.e., symptoms of job stress) and neurocognitive functioning based on the information from the video-recording including facial emotion expression, voice prosody, and other factors captured on video. In more technical terms, we apply Transfer Learning for feature extraction from raw video and audio recordings to develop digital biomarkers in order to test whether digital biomarkers can reliably predict future burnout symptom scores. The predictive value of candidate features is evaluated using supervised machine learning using the well-established, validated, and reliable Maslach Burnout Inventory (MBI9) as an outcome to predict.
If you are registered as a volunteer, please log in to contact the study team/express interest in this study.