PREMUS 2025: 12th International Scientific Conference on the Prevention of Work-Related Musculoskeletal Disorders
Electromyography of the forearm muscles during physical therapy tasks: evaluation and comparison of normalization methods
2German Social Accident Insurance (DGUV), Institution for the Health and Welfare Services (BGW), Hamburg, Germany
3Institute for Health Services Research in Dermatology and Nursing (IVDP), University Medical Center Hamburg-Eppendorf, Hamburg-Eppendorf, Germany
Text
Introduction: Physical therapists often experience work-related musculoskeletal disorders in the wrist and thumb [1]. It is important to quantify the load on these structures using measurement technologies such as electromyography (EMG), which can be incorporated into the CUELA risk assessment framework developed by the Institute for Occupational Safety and Health (IFA) of the German Social Accident Insurance (DGUV). Although EMG is widely used to assess muscular load, guidelines on forearm EMG normalization - critical for comparing individuals, trials, and muscles - are lacking. This work aimed to identify a normalization method for forearm EMG, including the thumb muscles, specifically tailored for field studies with physical therapists. The project is funded by the German Social Accident Insurance for the Health and Welfare Services (BGW).
Methods: 15 healthy physical therapists, equipped with seven EMG sensors per forearm, performed a series of reference contractions for normalization purposes using a force gauge dynamometer. These included five maximal voluntary contractions (MVCs) across four distinct grip positions (cylinder, pincer, fingertip, and lateral grip) and four submaximal reference voluntary contractions (RVCs) in the cylinder grip at different force levels (25%, 50%, 75% of MVC, and 100N). The normalization methods were evaluated regarding
- the ability of the MVCs to elicit true maximal activation in the investigated muscles,
- the intra-session test-retest reliability of the reference contractions, and
- the practical suitability for their use in field studies.
Results: On average, the MVC in the cylinder grip generated the highest EMG levels across all sensors and subjects, exceeding 90% of the maximum EMG recorded during any of the five MVCs. Performing the MVC in the pincer grip in addition, and taking the highest value from both, further increased the average EMG levels across all subjects for the thumb muscles (right: from 93.9% to 99.5%; left: from 93.2% to 97.4%). All reference contractions showed moderate to good reliability (intraclass correlation coefficients = 0.69-0.94; % standard error of measurement = 9.2-32.3%). When comparing the reliability of RVCs and MVCs in the cylinder grip, neither method demonstrated clear superiority over the other.
Discussion: Regarding the practical suitability for the use in field studies MVCs and RVCs at 100N were found to be comparable, as the complexity, time, and equipment requirements for these normalization methods were similar. However, as opposed to RVCs, normalizing to MVCs allows muscular load to be interpreted relative to a subject’s maximal capacity. Therefore, normalization to MVCs in the cylinder grip combined with the pincer grip for thumb muscles is preferred, while normalization to RVCs at 100N is a suitable alternative for individuals unable to perform maximal contractions due to pain.
Conclusion: The results provide a scientific basis for selecting an appropriate forearm EMG normalization method in field studies with physical therapists, contributing to the development of a sensor-based approach for measuring wrist/thumb load.
