BurnoutWords - Detecting Burnout for a Clinical Setting

Abstract

Burnout, a syndrome conceptualized as resulting from major workplace stress that has not been successfully managed, is a major problem of today's society, in particular in crisis times such as a global pandemic situation. Burnout detection is hard, because the symptoms often overlap with other diseases and syndromes. Typical clinical approaches are using inventories to assess burnout for their patients, even though free-text approaches are considered promising. In research of natural language processing (NLP) applied to mental health, often data from social media is used and not real patient data, which leads to some limitations for the application in clinical use cases. In this paper, we fill the gap and provide a dataset using extracts from interviews with burnout patients containing 216 records. We train a support vector machine (SVM) classifier to detect burnout in text snippets with an accuracy of around 80%, which is clearly higher than the random baseline of our setup. This provides the foundation for a next generation of clinical methods based on NLP.

BurnoutEnsemble: Augmented Intelligence to Detect Indications for Burnout in Clinical Psychology

“…A first attempt of working with clinical data to detect burnout has shown promising results. By presenting a dataset from real-world burnout patient data, Nath and Kurpicz-Briki ( 2021 ) managed to go beyond typical burnout detection approaches, which usually includes the use of inventories with scaling questions and worked on applying NLP to mental health. The dataset consisted of data extracted from German-language interviews with burnout patients, a control group, and experts.…”

“…To measure burnout risk, the metric is based on low valence and dominance and high arousal (Mäntylä et al, 2016 ). In other work, a first attempt to detect burnout based on patient and expert interviews in the German language were done; it was found that a combination of NLP and machine learning techniques in this field leads to promising results (Nath and Kurpicz-Briki, 2021 ).…”

Burnout and Depression Detection Using Affective Word List Ratings

“…Data set: An extended version of the data set from [9] consisting of texts of the following three classes was used: (i) Burnout (ii) Depression (online testimonials, transcripts of documentaries, online forums) and (iii) Control. Each category contains anonymized texts based on publicly available data originating from or having a strong relation to individuals suffering from burnout, depression or none of them (control group).…”

Predicting physical and mental health status through interview-based evaluation of work stress: initial attempts to standardize the interviewing method

“…44 Similarly, attempts have been made to detect burnout by analyzing the subject's speech content extracted from interview records. 45 The same potential for automation exists for stress assessment interviews. By collecting a large amount of highquality data in a standardized manner, it is possible to automate some aspects of the procedure, thereby reducing the burden on occupational health staff.…”