Using Explainable Supervised Machine Learning to Predict Burnout in Healthcare Professionals Karthik ADAPA a,b,1 , Malvika PILLAI a , Meagan FOSTER a,b , Nadia CHARGUIA c and Lukasz MAZUR a,b a Carolina Health Informatics Program, University of North Carolina (UNC), Chapel Hill, USA b Division of Healthcare Engineering, Department of Radiation Oncology, School of Medicine, UNC, Chapel Hill, USA c Department of Psychiatry, School of Medicine, UNC, Chapel Hill, USA Abstract. Burnout in healthcare professionals (HCPs) is a multi-factorial problem. There are limited studies utilizing machine learning approaches to predict HCPs’ burnout during the COVID-19 pandemic. A survey consisting of demographic characteristics and work system factors was administered to 450 HCPs during the pandemic (participation rate: 59.3%). The highest performing machine learning model had an area under the receiver operating curve of 0.81. The eight key features that best predicted burnout are excessive workload, inadequate staffing, administrative burden, professional relationships, organizational culture, values and expectations, intrinsic motivation, and work-life integration. These findings provide evidence for resource allocation and implementation of interventions to reduce HCPs’ burnout and improve the quality of care. Keywords. burnout, healthcare professionals, supervised machine learning 1. Introduction Burnout is an occupational hazard characterized by emotional exhaustion, depersonalization, and diminished personal achievement. Before the COVID-19 pandemic, 20-40% of healthcare professionals (HCPs) reported severe burnout [1]. The COVID-19 pandemic has further increased HCPs’ burnout to levels that pose a threat to maintaining a functioning healthcare workforce [2]. Burnout in HCPs can contribute to low quality of care, impair cognitive processes and lead to patient safety issues including patient harm [3]. Thus, there is an urgent need to examine the key factors contributing to HCPs’ burnout during the COVID-19 pandemic. HCPs’ burnout is a complex multi-factorial problem that is often affected by several non- linear factors. The US National Academy of Medicine (NAM) proposed a systems-based framework and identified evidence-based work system factors that contribute to HCPs’ burnout [4]. These factors are also further mediated by individual characteristics such as gender, age, and race. However, limited studies have utilized this theoretical model in 1 Corresponding Author: Karthik ADAPA; E-mail: karthikk@live.unc.edu. Challenges of Trustable AI and Added-Value on Health B. Séroussi et al. (Eds.) © 2022 European Federation for Medical Informatics (EFMI) and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). doi:10.3233/SHTI220396 58