The Effect of Online Reviews on Physician Demand: A Structural Model of Patient Choice

Social media platforms for healthcare services are changing how patients choose physicians. The digitization of healthcare reviews has been providing additional information to patients when choosing their physicians. On the other hand, the growing online information introduces more uncertainty among providers regarding the expected future demand and how different service features can affect patient decisions. In this paper, we derive various service-quality proxies from online reviews and show that leveraging textual information can derive useful operational measures to better understand patient choices. To do so, we study a unique data set from one of the leading appointment-booking websites in the US. We derive from the text reviews the seven most frequently mentioned topics among patients, namely, bedside manner, diagnosis accuracy, waiting time, service time, insurance process, physician knowledge, and office environment; and then incorporate these service features into a random-coefficient choice model to quantify the economic values of these service-quality proxies. By introducing quality proxies from text reviews, we find the predictive power of patient choice increases significantly, for example, a 6% to 12% improvement measured by MSE for both in-sample and out-of-sample tests. In addition, our estimation results indicate that contextual description may better characterize users' perceived quality than numerical ratings on the same service feature. Broadly speaking, this paper shows how to incorporate textual information into an econometric model to understand patient choice in healthcare delivery. Our interdisciplinary approach provides a framework that combines machine learning and structural modeling techniques to advance the literature in empirical operations management, information systems, and marketing.