Predictive modeling of presenteeism among radiographers: a secondary analysis of comprehensive data using Bayesian neural network

This study aimed to develop a predictive model for presenteeism among radiographers, integrating socio-demographic factors, work-related risks and musculoskeletal health. The prevalence of presenteeism was examined utilizing an already available dataset of 165 radiographers who completed an online survey via the ergonomic assessment of radiographers questionnaire. Of the respondents, 124 (75.2%) reported musculoskeletal dysfunction in the past 12 months, with 71 (43%) experiencing presenteeism. Binary logistic regression identified significant predictors: age (odds ratio [OR] 0.934, p = 0.032), weight (OR 0.830, p = 0.029), female gender (OR 0.226, p = 0.009), leave per week (OR 0.275, p = 0.036), static working posture (OR 7.867, p = 0.036) and musculoskeletal pain in the last 12 months (OR 108.938, p < 0.001) with area under the receiver operating characteristic curve (AUROC) of 0.86. The Bayesian neural network model also exhibited an AUROC of 0.74, indicating strong discriminatory power. This study underscores the association of personal, ergonomic risk and musculoskeletal factors with presenteeism among radiographers. Trial registration: Clinical Trials Registry–India identifier: CTRI/2021/09/036992.