This study presents a machine learning approach for modeling and predicting COVID-19 patient outcomes in the Republic North Macedonia based on demographic characteristics. A decision tree-based classification model has been developed using algorithms implemented in WEKA environment. The model constructs the decision tree by selecting random subsets of features at each node and determines the optimal split points based on information gain. The results indicate that age is the most significant predictor of patient outcome, serving as the root node and primary splitting attributes in the decision tree. Individuals below 60 years of age are classified as recovered, while higher age groups exhibit an increased probability of death. The gender attribute appears as a secondary factor, contributing to more refined predictions within specific age intervals. In addition to its predictive capability, the model offers a high level of interpretability, enabling clear visualization of decision rules and facilitating understanding of the relationship between demographic factors and disease outcomes. These findings highlight the importance of age- based risk stratification and demonstrate the potential of machine learning models as decision support tools in public health management.