Construction and evaluation of a risk prediction model for work-related musculoskeletal disorders among construction workers at a hydropower station on the Qinghai-Tibet Plateau: a cross-sectional study.

INTRODUCTION: The use of outdated building models and inefficient management practices continue to pose serious risks to the occupational health of workers in China's rapidly expanding construction sector. We identified key risk factors for work-related musculoskeletal disorders (WRMSDs) among hydropower station construction workers in the Qinghai-Tibet Plateau and developed a risk prediction model.

METHODS: This cross-sectional survey of 325 workers was conducted in December 2025. Least absolute shrinkage and selection operator regression was used for variable selection and multivariate logistic regression was employed to construct a prediction model visualized as a nomogram. Model performance was validated using receiver operating characteristic, calibration, and decision curve analysis.

RESULTS: The prevalence of neck, shoulder, and lower back WRMSDs was 31.1% (101/325). Five key predictors were identified: household registration type, chronic disease history, sleep duration, work type, and workplace temperature. The model demonstrated good discrimination (area under the curve = 0.789, 95% confidence interval 0.724-0.853) and calibration (p > 0.05). Urban household registration (odds ratio OR = 2.24, p = 0.036), electromechanical installation work (OR = 3.62, p = 0.011), other occupations (OR = 5.56, p < 0.001), chronic disease history (OR = 6.02, p = 0.018), and cold workplace (OR = 2.47, p = 0.006) were risk factors, whereas longer sleep duration was a protective factor (OR = 0.80, p = 0.049).

DISCUSSION: In this population, WRMSDs were influenced by demographic, health, and environmental factors. The developed model enables early identification of high-risk workers and supports targeted occupational health interventions.