This study investigated the information behaviour of people infected with, or affected by HIV/AIDS (PI/A HIV/AIDS). It sought to generate in-depth knowledge of the real world 'lived information experiences' of PI/A HIV/AIDS, with the aim of informing information management practice, strategy and policy to better contribute to the fight against HIV/AIDS. Ethnography was selected as the methodological approach since it is particularly well suited to unearthing complex phenomena, in marginalised social settings, among hard to reach populations. It has the potential to uncover what might otherwise remain hidden aspects of social experience particularly in settings characterized by stigma and discrimination. The sample studied consisted of 40 PI/A HIV/AIDS in the UK, recruited at an HIV support centre where the researcher worked as a part-time volunteer. It included a range of age groups, disease stages, sexual orientations, and educational and ethnic backgrounds. The interviews were analysed using Spradley's Domain Analysis. The research has contributed to knowledge by revealing a number of distinctive information behaviours, including not only information seeking and sharing, but also covert and hysterical information seeking, destroying and hiding information, which appeared to be strategies for coping with a range of internal and external factors: reactions to diagnosis, stigma, and the need to educate and inform both self and others. The research has highlighted the complex and dynamic shifting nature of HIV/AIDS-related information behaviour of PI/A HIV/AIDS. A model which attempts to describe patterns of HIV/AIDS-related information behaviour and explains why such patterns could develop in the HIV/AIDS-related information context has been developed based on the lived information experiences of PI/ A HIV/AIDS. The model moves beyond relatively reductionist and simplistic representations of information behaviour by addressing finer details of the complexities of human information behaviour, particularly in life-threatening and stigmatising health-related information interactions. This could potentially illuminate ways in which information services could be responsive to the needs of PI/A HIV/AIDS, and provide a basis for proposing recommendations for information management in health contexts. Thus, in addition to paying attention to how and when people relate to information, the model has generated rich insights about complexities, subtleties, challenges and dilemmas experienced by people faced by the life-threatening and stigmatising condition such as HIV/AIDS. In addition to identifying information behaviours within the HIV/AIDS context, the model has drawn our attention to forms of information seeking which have received only scant attention in the library and information science research literature, such as covert and hysterical information seeking. The model draws on theoretical propositions from Stress Appraisal and Coping Theory and Stigma Management Theory to explain HIV/AIDS-related information behaviours. The model acknowledges the role of emotional as well as rational factors in determining information behaviour. Recommendations are made relating to the need for information professionals and information service designers to understand information users at deeper levels if their information services and products are to contribute to the prevention and management of HIV/AIDS.