Abstract: Objective To conduct a scoping review of the research on risk prediction models of work-related musculoskeletal disorders at home and abroad. Methods The database was systematically searched in both English and Chinese from its inception to August 20,2024. Two reviewers independently screened and extracted data according to the critical appraisal and data extraction for systematic reviews of prediction modelling studies(CHARMS) checklist,and assessed model bias and applicability with the prediction model risk of bias assessment tool(PROBAST). Results A total of 13 papers were included,most of the research subjects were nurses,most of the model construction methods were logistic regression models,the model display was mainly based on the visual display of nomograms,and the three predictors with the highest frequency were length of service,posture and position,and repetitive action. Conclusion None of the existing prediction models for work-related musculoskeletal disorders fully meet the requirements of model quality and applicability. At present,the standard model should adopt the specifications,such as transparent reporting of a multivariable prediction model for individual prognosis or diagnosis(TRIPOD),which can improve the quality and transparency of the model.

Key words: Work-related musculoskeletal disorders, Predictive models, Risk assessment, Scoping review