People diagnosed with chronic illnesses are increasingly turning to the Internet to search for information about their health. In some cases, these individuals also disclose personal health information. In fact, these two information behaviors are often linked. This is particularly true in online support groups. However, we do not currently understand how and why people diagnosed with chronic diseases undergo these activities – although they are becoming increasingly common as the Internet becomes more ingrained in everyday life. This dissertation uses constructivist grounded theory to examine health information seeking online, personal health information disclosure, and the relationship between these processes. Constructivist grounded theory is an ideal research method for examining phenomena, behaviors, and processes that have not yet been fully explored, as is the case in this study. In this study, twelve participants diagnosed with chronic kidney disease were recruited in three different online support groups for chronic kidney disease. Chronic kidney disease was chosen as the illness context in this study as it is a non-stigmatized, incurable, life-long condition that requires patients to manage their treatment over time. Two telephone interviews were conducted with each participant and their comments to online support groups were collected. Data were collected and analyzed using inductive analysis, the constant comparative method, memoing, and theoretical sampling strategies. Trustworthiness of the analysis was obtained using multiple methods, including peer de-briefing and member checking. A model of information seeking, personal health information disclosure, and similarity assessment in online support groups was derived from the analysis. The model highlights one of the central contributions that this study makes to our current understanding of information behavior: similarity assessment, or the process of finding people who are similar in online support groups both as human sources of information and as people with whom to disclose. The assessment of similarity changes over time as experience and knowledge evolve. The findings presented in this dissertation add to our theoretical understanding of information behavior; they also intersect with theories from other disciplines, including communications and nursing. They also have practical implications for healthcare providers and designers of information systems.