How do patients' perceptions and doctors' images impact patient decisions? Deconstructing online physician selection using multimodal data

Abstract

In the post-pandemic era, medical resources are uneven, and access to healthcare is complicated. Online medical platforms have become a solution to bridge the information gap and reduce hospital pressure. This study uses the stereotype content model and signaling theory to explore the impact of patient perception of patient decision making (PDM) on online medical service platforms. Also, it tests the moderating effect of physician image. We collected information on 12,890 physicians and 746,981 patient reviews from online medical platforms in China. Unsupervised machine learning was used to construct a topic model to extract patients' perceptions of physicians' competence and warmth. Meanwhile, the facial features of physicians, such as age, smile, and glasses, are recognized by convolutional neural networks. Finally, the influence of PDM concern on decision-making and the moderating effect of physician image were analyzed by multiple linear regression. The results of the study showed that (1) patients' perceptions of physicians' competence and warmth had a positive effect on decision-making; (2) physicians' age and wearing glasses enhanced the positive effect of perception on decision-making; and (3) however, physicians' smiles weakened the positive effect of perception on decision-making. This study provides new insights into patients' online physician selection, guides the construction and promotion of medical service platforms, and provides an effective avenue of exploration to alleviate the problem of uneven distribution of offline medical resources.

Keywords

• Convolutional neural network
• Online medical service platform
• Patients' decision-making
• Physicians' online image
• Structural topic model
• Stereotype content model