Identifying the Key Predictors of Occupational Fatigue among Long-Distance Truck Drivers in East Africa: A LASSO-Regularized Regression Approach

ObjectivesThis study aimed to determine the prevalence of occupational fatigue and identify its primary risk factors among long-distance truck drivers operating along the key Kenya-Uganda transport corridor, a vital artery for regional commerce where comprehensive data has been limited. MethodsA cross-sectional analytical study was conducted with 207 exclusively male long-distance truck drivers at the Busia and Malaba border points. Participants completed structured questionnaires capturing demographics, work patterns, sleep habits, and stimulant use. Fatigue was assessed using the Chalder Fatigue Scale. Data were analyzed in RStudio, using LASSO regression with 10-fold cross-validation for predictor selection to address multicollinearity. Selected variables were used in a multivariable logistic model to calculate adjusted odds ratios (aORs). Bootstrap validation assessed the models performance. ResultsThe overall prevalence of occupational fatigue was 51.7%. Key risk factors identified included high pressure to meet deadlines, the use of stimulants (e.g., caffeine, khat) to maintain alertness, and excessively long average shift lengths. The multivariable model demonstrated excellent and stable performance, with a mean Area Under the ROC Curve (AUC) of 0.987 across 1,000 bootstrap samples. ConclusionsOccupational fatigue is highly prevalent among long-distance truck drivers in this region, driven largely by organizational factors. The findings highlight an urgent need for multimodal interventions, including enforceable regulations on driving hours, targeted driver education, and improved scheduling practices by transport companies to safeguard driver well-being and public road safety. Key MessagesO_ST_ABSWhat is already known on this topicC_ST_ABSTruck-driver fatigue is a global problem, but evidence from East African corridors has been minimal. What this study addsFatigue prevalence on the Kenya-Uganda corridor is high (51.7%). LASSO analysis highlights key predictors: deadline pressure, stimulant use, long shifts, and poor rest. How this study might affect research, practice or policyFindings support driving-hour regulations, improved scheduling by transport companies, and targeted public health messaging on stimulant risks.