Social media platforms for healthcare services are changing how patients choose doctors. The digitization of healthcare reviews has enabled patients to thoroughly evaluate doctors before booking an appointment, and has increased the transparency of the relationship between patients and doctors. In this paper, we wish to derive the impact of online information on patient choice of outpatient care doctors. We are especially interested in how operational factors influence demand. To do so, we study a unique data set from one of the leading appointment booking websites in the United States, that contains online doctors' appointments made over a five-month period, along with other online information. We propose a random coefficient logit model to characterize consumer heterogeneity in doctor choices, taking into account both numeric and textual user-generated content with text mining techniques. Based on the estimation of the model, we infer the impact of various operational factors, such as location flexibility, online appointment book, service time, and waiting time, on demand. We proceed with counterfactual experiments, and simulate the impact of proposed policy changes. We then provide managerial insights derived from these experiments, and make suggestions to help increase doctors' demand. On a broader note, this paper illustrates how social media can be mined and incorporated into a demand estimation model to derive the impact of operational factors in healthcare delivery. Our interdisciplinary approach provides a framework that combines machine learning and structural modeling techniques with empirical operations management.