Text

Introduction: Artificial intelligence (AI) has emerged as a transformative technology of the 21^st century, with significant potential in the fields of health promotion and healthcare. To date, AI has been used in the areas of return to work and occupational safety and health, mainly to improve risk assessment or vocational rehabilitation through the implementation of AI algorithms. Only a limited number of studies are addressing workplace health prevention and promotion (WHPP) [1]. Musculoskeletal disorders are particularly underrepresented [1], despite being one of the major causes leading to sick leave [2]. This study addresses this gap by investigating the effectiveness of a personalized AI-based health assistant in reducing musculoskeletal complaints and improving health-related behaviors among office workers.

Methods: A randomized controlled trial is conducted with full-time office workers suffering from musculoskeletal complaints. The intervention group uses the AI-based health assistant during working hours for 12 weeks. The assistant assesses sitting behavior, provides exercise recommendations, and sends multiple prompts to encourage health behavior changes (e.g., posture change, sedentary breaks). The control group receives no intervention. The primary outcome is change in musculoskeletal complaints, measured for current, average, and maximum complaints using an 11-point numerical rating scale ranging from no complaints to the worst imaginable complaints. Secondary outcomes include self-perceived changes in health-related behaviors. Data are collected via online questionnaires at baseline, after 2, 4, 8, 12 weeks, and 6 weeks post-intervention.

Results: Recruitment is completed with 99 eligible participants (60% female; mean age 40.02 ± 10.5 years). Results of the statistical analyses regarding the effectiveness of the intervention to will be presented at the congress. We hypothesize that the intervention group will show a significant reduction in musculoskeletal complaints compared to the control group. Additionally, the study will examine the subjective perception of health-related behavior change. We assume that the intervention group will demonstrate a significant self-perceived improvement in health-related behavior.

Discussion: The study will provide insights into whether a personalized AI-based health assistant effectively reduces musculoskeletal complaints and promotes health-related behavior change in the workplace.

Conclusion: By focusing on musculoskeletal health, this study addresses a critical gap in current research and contributes evidence on the effectiveness of AI-based interventions in the field of WHPP.

Literatur