End-of-life discussions with noncurable lung cancer patients: A patients' and oncologists' qualitative study.

151 Background: Discussions concerning patients' (pts) preferences regarding end-of-life (EOL) care can improve the quality of care in non-curable-cancer-pts. Nevertheless having such discussions is difficult for health care providers and discrepancies exist regarding the frequency of such discussions.We conducted a single center qualitative study exploring oncologists' attitudes and practices toward EOL discussions and the feasibility of an intervention designed to explore and collect pts' preferences regarding EOL care. For oncologists, the intervention consisted in a semi structured interview and in a questionnaire concerning pts included. For pts, semi-structured-in-depth-interviews were conducted to assess their opinion about EOL discussions and the feasibility of exploring and collecting their preferences regarding EOL care in our practice setting. Resulting anxiety was measured.Oncologists' interviews (n=5), revealed that EOL discussions were rare, more likely to be initiated by pts and to arise lately in the evolution of the disease. Main barriers invoked were the difficulty of having such conversation, the limited time available, the risks of maleficence (mainly anxiety and loss of hope) and the uncertainty concerning expected benefits. Pts' interviews (n=10) revealed that most of them were willing to be part of medical decisions concerning their EOL care. Few preferences regarding EOL-care were actually collected but the resulting in-depth discussions help mentioning a great variety of unusual topics as well as emotional, cultural and spiritual issues. Anxiety as a side effect was uncommon. Interviewers and oncologists often overestimated pts' anxiety which is consistent with a projection of their own anxiety.Discussions concerning EOL care are uncommon in the setting studied and four main barriers were identified. Having these conversations was feasible and led to unexpected in-depth interviews concerning topics rarely addressed in our current practice. This intervention could help clarifying pts' preferences and identifying psychosocial issues that could trigger palliative care team referral.