User technology for AI-supported objective measurement of Work-related Quality of Life (MI-LQ)

Abstract Background Understanding and enhancing job satisfaction is crucial for improving Work-Related Quality of Life (WrQoL) in today’s office work environments. Our work uses multimodal data collection and analysis techniques to objectively assess WrQoL, thus mitigating respondent bias and survey fatigue. Leveraging machine learning (ML) and automated methods measuring WrQoL presents a promising approach in public health research, particularly concerning stress-related health outcomes. Methods Through a systematic literature search of 9,242 publications to identify WrQoL indicators, we selected over 100 publications for data extraction. A base model, including workload, commute, overtime, valence, and arousal was formulated. To amass a wide data set, Objective measurement tools (computers, smartphones, and wearables) complemented by questionnaire data. This data will inform ML algorithms to predict individual WrQoL indicators. Developed as an input/output data platform, an app-prototype will objectively assess WrQoL, facilitating a field study with office workers to refine the model. A mobile app prototype serves as a user interface and data relay for WrQoL metric analysis. Results Our research develops instruments for measuring WrQoL using smartphones, wearables, and specialized trackers for heart rate variability (HRV) analysis, alongside a software prototype for keyboard and mouse tracking for workload assessment of valence and arousal. Fitness trackers with photoplethysmography functionality are favored for HRV assessment. Conclusions The systematic search identified influences on WrQoL, translated into measurable indicators in a base model. Leveraging user tech for assessment and a ML and mobile app framework, this approach facilitates objective WrQoL determination. By presenting the relationship between digital service use and WrQoL indicators, this methodology promises new insights, informing stress and burnout prevention strategies within public health initiatives. Key messages • Progressive approach: ML-driven WrQoL assessment offers promising avenues for stress prevention in workplace health research. • Methodological Advancement: The novel framework for WrQoL assessment using digital tools contributes to effective workplace health promotion strategies, with potential for tailored interventions.