People’s activities on the internet have expanded from mainly retrieving information to communicating with each other in virtual settings. Thus, research investigating social interactions in online communities is becoming more and more important. However, the multi-faceted approaches of existing studies for the analysis of online communities make it difficult to combine the findings into a comprehensive understanding. This shows the need for holistic investigations of online communities. This thesis provides such a holistic approach by investigating a combination of different aspects of a selected online support community for older people. MOSuC (Model of Online Support Communities), a model describing the key aspects of online support communities was developed based on existing theories of computer-mediated communication (CMC) as well as theoretical perspectives on social support. Five studies were conducted, each addressing one of the different aspects of the case study community: (i) the message content, (ii) the conversations structure, (iii) the social network of related community members, (iv) the roles that online community members take on, and (vi) the taget population’s needs concerning the exchange of social support in online communities. The findings of these individual studies were then combined in context of MOSuC in order to provide a holistic description of the community. As a result, this thesis provides detailed insight into the characteristics of the case study community as well as the interplay and dependencies between different aspects of the community. Based on the integration of multiple studies, the thesis sheds light on two main issues: the characteristics of the individual aspects of the community as well as how these aspects are related to and affect each other. In addition to the findings of the studies, the thesis also contributes MOSuC, which serves both as a theoretical framework of the aspects of online support communities, as well as a practical tool for integrating the individual studies. In addition, the application, modification and integration of multiple methods in this thesis provide a novel methodological way for an integrative analysis of online support communities.