This paper addresses the challenges of classifying healthcare-related texts from online healthcare forums (OHFs). Given the unstructured nature of OHF posts, traditional classification algorithms face difficulties. Moreover, while advanced models like deep neural networks offer high predictive accuracy, their interpretability remains a challenge, which is crucial for healthcare applications. To address these issues, the paper proposes an effective and interpretable OHF post classification framework. The framework categorizes sentences into three classes: medication, symptom, and background. It uses an interpretable feature space with labeled sequential patterns, UMLS semantic types, and heuristic features. A forest-based model is employed for classification, and an interpretation method is developed to extract decision rules, providing insights into the information in the texts. The proposed framework's effectiveness is validated through experiments on real-world OHF data.