Using machine learning models to study the magnitude and associated factors of road traffic accidents among bus drivers in Marodijeh Region, Somaliland

Road traffic accidents (RTAs) represent a critical global health and development challenge, disproportionately affecting low-income countries. Despite the high fatality rates in Sub-Saharan Africa, there is limited research utilizing advanced predictive modeling to understand accident associated factors in Somaliland. This study aims to assess the magnitude of RTAs and identify risk factors among bus drivers in the Marodijeh region using both classical statistical methods and machine learning (ML) algorithms. A cross-sectional study was conducted involving 486 bus drivers selected via convenience sampling. Data were collected through face-to-face interviews using a structured questionnaire. The study employed a multivariable binary logistic regression model alongside five machine learning algorithms—Random Forest (RF), Decision Tree (DT), Gradient Boosting Machine (GBM), Support Vector Machine (SVM), and Logistic Regression—to predict accident occurrence and identify feature importance. The prevalence of RTAs among bus drivers in the past two years was 52.7% (95% CI: 48.2%–57.2%). The Random Forest model outperformed other models, achieving the highest accuracy (76.55%), sensitivity (73.91%), and Area Under the Curve (AUC). Key factors identified included mobile phone usage while driving (AOR = 3.68), lack of vehicle fitness certification (AOR = 0.37), vehicle service history, and khat chewing (AOR = 1.92). Contrary to general trends, higher education levels and possession of a driving license were associated with increased accident odds in this specific context. The study reveals a high magnitude of traffic accidents among bus drivers in the region. The superior performance of the Random Forest model highlights the potential of machine learning in developing targeted road safety interventions. To align with Sustainable Development Goal 3.6, policy recommendations include stricter enforcement of bans on phone usage and khat chewing while driving, alongside mandatory regular vehicle inspections.