Ethnography (and participant observation are increasingly used in palliative care research (1). In this presentation we want to demonstrate the unique contribution of an ethnographic approach to research in palliative care by describing how this approach enabled us to identify the core research question of our research on "false optimism about recovery" of patients with small cell lungcancer (2) as well to identify the mechanisms by which this optimism was "produced" and maintained. Our research topic was "discovered" in an ethnograhic study of euthanasia in a lung diseases department of a university hospital. Before this "discovery", this optimism was not discussed in literature, nor had practioners mentioned it as a significant problem in their patients. The fact that the phenomenon was considered "non-reportable" was both a reason for its existence and for its absence in the clinical and research literature (participants did not deem it problematic). Next we discuss how mechanisms (such as ways of not "hearing bad news") that contributed to the phenomenon of "false optimism" could only be "discovered" and understood when the researcher experienced them herself in her communication with patients. She also experienced herself "naturally" reverted to methods which she had seen doctors and nurses using to avoid painful confrontations with patients and their relatives. These examples demonstrate how participant observation enables reseachers to "discover" and understand relevant phenomena. It is our contention that such data can rarely (or only with more difficulty) be obtained through interviews. 1. Jane Seymour. Critical moments: death and dying in intensive care. Buckinham OUP. 2. Anne-Mei The. Palliative care and communication; experiences in the clinic. Buckingham: OUP. The, A-M, A Hak, GH Koeter, G van der Wal. Collusion in doctor-patient communication: an ethnographic study. BMJ 321:1376-81.