<![CDATA[Low back pain (LBP) is a common musculoskeletal disorder among office workers and reduces work productivity. Studies that integrate workstation risk and sitting behavior into a prediction model for office workers in the electronics manufacturing industry remain limited. This study aimed to develop a prediction model of LBP based on workstation risk and sitting behavior among office workers at PT. X Manufacturing Elektronik in Batam. An analytical observational study with a cross-sectional design was conducted among 122 office workers selected through total sampling. Workstation risk was assessed using the Rapid Office Strain Assessment (ROSA), sitting behavior (daily sitting duration and microbreak habits) was measured by questionnaire, and LBP was identified using the Nordic Musculoskeletal Questionnaire (NMQ) based on symptoms during the previous seven days. Data were analyzed using the chi-square test and binary logistic regression with backward elimination. The prevalence of LBP was 45.1%. Workstation risk, sitting duration, and microbreak habits were significantly associated with LBP (p < 0.001). In the final model, irregular microbreaks (OR = 3.90; 95% CI: 1.32–11.49), sitting for more than 6 hours per day (OR = 3.39; 95% CI: 1.15–9.97), and high workstation risk (OR = 2.97; 95% CI: 1.01–8.71) were independent predictors. The model showed a Nagelkerke R² of 0.371, a classification accuracy of 74.6%, and an area under the curve (AUC) of 0.784. The model may serve as a basis for initial LBP screening and risk-based ergonomic programs; however, the cross-sectional design and limited sample size call for validation in other populations.]]>
