Analyzing Patient Decision Making in Online Health Communities

In recent years, many users join online health communities (OHC) to obtain information and seek social support. One of the most important type of support that a patient needs is to obtain suggestions and information when they make decisions in their diagnosis and treatments. To understand patient decision making processes, we propose to identify the threads on OHC discussion forum that are about patient decision making, and then analyze the questions that patients have. We use deep learning based model to identify such threads. Experiment results show that the proposed methods achieve good performance in precision, recall, F1 score, accuracy, and AUC. Then we leverage topic modeling techniques to analyze the questions that patients expressed in those threads to get a better understanding of patient decision making.