¹The National Research Center for the Working Environment, Copenhagen, Denmark
²University of Southern Denmark, Odense, Denmark
³KTH Royal Institute of Technology, Stockholm, Sweden

Text

Introduction: Musculoskeletal disorders (MSD) are a leading cause of work-related disability, reduced productivity, and sickness absence across Europe, with estimated productivity losses of up to 2% of GDP. Neck-shoulder pain (NSP) is among the most prevalent and disabling MSDs in physically demanding jobs. Accelerometers are widely used to objectively measure physical exposures. However, current metrics, such as time in angular zones, movement amplitude, and velocity, often lack interpretability for ergonomists, who rely on observational workplace assessments. To address this, a novel, expert-based method for identifying arm movements from accelerometry has been developed. The aim of this study is to investigate whether the accumulated number of arm movements during work is associated with the risk of NSP one year later.

Methods: We applied an expert-based arm movement definition, developed through a Delphi study with 17 European researchers, balancing biomechanical relevance with practical observability. This definition will be applied to accelerometry data collected from the dominant upper arm across five workdays. A custom algorithm detects arm movements using expert-agreed thresholds. An arm movement is defined as a ≥20° vertical shift in arm direction. Identified movements are categorized by velocity (fast/slow), dynamics (static/dynamic), amplitude (high/low), elevation zones (0–30°, 30–60°, 60–90°, >90°), and frequency (per 60-minute interval). Monthly self-reports of NSP were collected for one year. At the time of the conference, we will present associations between arm movement frequency and the risk of reporting NSP after one year, analyzed using generalized linear mixed models adjusted for age, sex, BMI, baseline pain, sector, and self-rated health.

Results: The sample (N = 1,087) had a mean age of 45.2 years, 51.6% male, BMI of 27.2 kg/m², job seniority of 13 months, and weekly working hours of 37.5. Baseline pain intensity averaged 3.4 (0 = none, 10 = worst); self-rated health was 2.2 (1 = very good, 5 = very poor). Sector distribution: manufacturing (71.4%), cleaning (19.8%), and transportation (8.8%). During upright work, participants spent on average 94 minutes with arm elevation >30°, 17 minutes with >60°, and 3 minutes with >90°. Variation in time spent with arm elevation among workers enables statistical analyses of associations with NSP.

Discussion: This study is among the first to apply an expert-defined concept of arm movement to real-world occupational data using wearable sensors. Although it was designed to reflect clinically relevant movement patterns, its predictive validity is currently being evaluated. In addition, reducing complex arm movements to thresholds and categories may oversimplify real-world behavior. Future research may explore the method’s applicability across different work contexts, validate it against other outcomes, and consider integrating complementary physiological or contextual measures.

Conclusion: If the number of arm movements is found to be associated with one-year risk of NSP, the proposed method may offer an interpretable and actionable alternative to conventional arm exposure metrics, supporting ergonomic risk assessments and interventions in manual work.