BACKGROUND

Online healthcare has become an important part of healthcare services. Patients can access unlimited information reflecting a doctor's service quality in online health communities (OHCs). This information can reduce information asymmetry between patients and doctors, and further promote patient decisions. In this study, we focus on the patient's decision to switch from a doctor's online medical service to the same doctor's offline service.

OBJECTIVES

We classify OHC information into objective information and subjective information. Following this, we investigate how these two types of information influence patient decisions to switch from a doctor's online to offline medical service, and explore the nature of interaction between the two types of influence.

METHODS

We collected data from a leading OHC in China and constructed a longitudinal dataset to examine our research questions. To control for doctor heterogeneity and common trends across time, we leveraged fixed effects at the doctor level and monthly fixed effects in our models, respectively.

RESULTS

Our findings demonstrate a strong relationship between OHC information and patient decisions to switch from a doctor's online to offline medical service. Patients are more likely to switch from online to offline medical services with doctors whose subjective and objective information on their homepages indicate a higher quality (β = 0.001, p < 0.001; α = 0.019, p < 0.002). We also observe that patients prefer doctors who do not exhibit objective information that indicates low-quality service over those who show such information on their homepages, even though the chosen doctor provides a lower-quality service (α = -0.032, p < 0.006). Further, we find that the influence of subjective information on patient decisions crowds out the influence of objective information.

CONCLUSION

This study investigates the role of subjective and objective information on patient decisions to switch from a doctor's online to offline medical service and the interaction effect between these two types of information. This study provides valuable theoretical and practical implications.