Predicting User Participation and Detecting User Role Diffusion in Online Health Communities

Nowadays, healthcare issues draw people's attention globally. According to the report by the Pew Research Center [1], 35% of U.S. Adults have gone online particularly for information related to medical conditions. Besides getting information from healthcare professionals and friends, 24% of adults also sought information or support from peers who have the same health condition. A major venue where people find such peers is Online Health Communities (OHCs), such as "Patientslikeme.com". Compared with traditional health-related websites that only allow users to retrieve information, OHCs increased members' ability to interact with peers facing similar health problems and, as a result, better meet their needs for social support. Literatures on social support suggest that OHCs mainly feature three types of social support: informational support, emotional support, and companionship (a.k.a., Network support) [2]. As social support is a pillar of OHCs, a natural question to ask would be: when it comes to users' participations, are a user's online activities in different types of social support related to her/his participation in an OHC? If so, can we predict whether a user will churn from an OHC based on these social support activities? Despite the large amount of research on social support in OHCs, few studies have answered this question systematically by examining users' seeking, receiving, and provision of various types of social support from large-scale datasets. In 2012, Wang [3] suggested that receiving more emotional support is associated with users' longer stay in an OHC. However, the types of social support investigated were limited and only the receiving of support was considered, while that providing social support is also important and beneficial. Analyzing large-scale data from an OHC, we combined various data analytics techniques, including text mining, survival analysis, and predictive modelling. We found that receiving more emotional support or contributing more companionship are positively correlated with users' longer involvement in OHCs. Our research described as well as predicted users' participation in OHCs from the perspective of online social support. At the same time, we identified several roles of users in an OHC and suggested that overall those who only seek social support are more likely to leave (i.e., "churn" from) an OHC. By analyzing their monthly behaviors in social support, we found that users' roles evolve over time. Moreover, we also found evidence for the diffusion of roles in the OHC social network -- a user tends to adopt the role that many of her network neighbors take. This is the first study to show that user roles related to social support are in fact "contagious" via social ties. The outcome of this research has implications for building and sustaining an active OHC.