<![CDATA[Introduction: Workers in informal industries are routinely exposed to physical, environmental, and ergonomic stressors that may be associated with fatigue and musculoskeletal strain. This study examined the cross-sectional associations of repetitive-task risk, sleep quality, body mass index, and selected individual characteristics with work fatigue among fish-smoking workers in Semarang, Indonesia. Methods: Using a cross-sectional design, data were collected from 64 workers through structured questionnaires, observational assessments, and anthropometric measurements. Fatigue levels were measured using the Indonesian version of the Japanese Industrial Fatigue Research Committee (JIFRC) Fatigue Questionnaire, repetitive-task risk was assessed using the Assessment of Repetitive Tasks (ART) method, and sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI), which evaluates subjective sleep quality over the preceding month. Results: Moderate fatigue was reported by 59.4% of participants. In the bivariate analysis, left-hand repetitive-task risk was associated with fatigue (p = 0.037), whereas right-hand repetitive-task risk and the dichotomised sleep-quality category were not associated with fatigue. In the adjusted linear regression model, the left-hand association was attenuated (B = 0.028, p = 0.839), while BMI remained associated with the coded fatigue outcome (B = 0.181, 95% CI 0.047 to 0.316, p = 0.009). The overall regression model explained 21.3% of the variation in fatigue (R² = 0.213; adjusted R² = 0.115) and did not reach the conventional threshold for statistical significance (F (7, 56) = 2.166, p = 0.051). These estimates should be interpreted cautiously because of the small sample and imbalanced categories. Conclusion: Among workers in the participating fish-smoking centres, BMI was associated with the coded fatigue outcome after adjustment, while the unadjusted association involving left-hand repetitive-task risk was not retained in the adjusted model. Because the overall regression model was marginally non-significant, the individual adjusted associations should be interpreted cautiously. The findings identify anthropometric assessment, fatigue screening, and further task-specific ergonomic evaluation as priorities for future health-promotion research. Longitudinal and intervention studies are needed before conclusions can be drawn regarding causality, intervention effectiveness, or productivity.]]>
