Background and aim The frequency and indications of use of drugs with sedative effect, for example, midazolam, for dying patients has been widely explored. Some authors have sought empirical evidence to argue a rationale that secures an ethical basis for sedation use by distinguishing it from euthanasia. A palliative care philosophy, that simultaneously promotes a symptom-control focus and a rejection of intentions to hasten death, underpins recent frameworks guiding sedative use. The authors present here empirical evidence to analyse the professional and clinical contexts influencing doctors and nurses when making ethical decisions to treat dying patients with sedatives.

Methods Ethnographic methods in each of three UK palliative care units (PCUs): participant-observation of inpatient ward care for approximately 100–150 h, writing field notes after observing staff managing dying patients; 10–12 recorded interviews and two focus groups with doctors and nurses. Use of NVivo9 to thematically analyse transcribed data, and support cross-comparison of observations between PCUs. Data collection and analysis will be complete by January 2012, and data deposited with the UK Data Archive thereafter.

Results and conclusions The main ethical justification for sedative use is that a dying patient's condition requires this. The degree of latitude given to nurses to administer doses of sedatives differs between PCUs; appearing related to the seniority of available nurses and the degree of medical control over nurses' practice generally. Staff actively manage relatives' impressions about sedation, often aiming to refute the intention of hastening death. Sedative drugs appear unique, compared to other end-of-life care drugs, facilitating the ‘emotional labour’ nurses must do to maintain a calm and orderly ward atmosphere and assisting in the production of an apparently ‘natural’ death. This in turn suggests that a dying patient's condition is not the only consideration that is relevant when choosing to use sedatives.