An Empirical Study of Online Supports among Patients

This paper investigates the helpfulness of an online healthcare community to patients’ health conditions.We propose an inhomogeneous Partially Observed Markov Decision Process (POMDP) model to examine the latent health outcomes of online health community members. The transition between different health states is modeled as a probability function which incorporates different forms of social supports that patients receive via online communication, and other factors that would impact patients’ online behaviors. We find that patients gain benefits from learning from others, and the participation in the online community helps them to improve their health conditions and better engage in their disease self-management processes. Our results also reveal the effect of various forms of social supports on dynamic evolution of patients’ health conditions. Informational support on average contributes the most to helping patients move to a better health state. Peer recognition and self-identification are also found to have significant impact on the transitions between patients’ health states. Finally, we demonstrate that our POMDP model provides a very accurate method to recover missing or unavailable information on patient health conditions.