RESUMO
OBJECTIVE: To determine the prevalence, epidemiology and registration status of palliative sedation (PS) prevalence in a teaching hospital, and to establish areas for improvement. METHODS: A descriptive retrospective analysis was designed using the records from cancer patients who died between October and December 2010. The variables included were: epidemiological, inpatient unit, refractory symptom, drugs and dosages, and patient participation in the decision making process. The qualitative analysis followed a Delphi process: each participant received the overall performance of the group referred to as mean, median, 25th and 75th percentile. Items selected were those in which there was total or a high consensus. RESULTS: A total of 53 deaths were identified. Just over half (51.92%) received PS. The mean age was 67.46 and 64% were males. The most frequent diagnosis was lung cancer (32.14%). Fifteen of the patient patients were in the Oncology ward, 7 in Hematology, and 4 at the Emergency Department. The PC team took part in 14 of the sedations performed. A refractory symptom was identified in 20. There were 11 cases of dyspnea and 5 cases of delirium. The mean time between admission and PS was 9.5 days. The mean duration of PS was 1.2 days, with a mean number of 2.6 drugs used. There were 20 informed consents which were all verbal. The mean time from last chemotherapy to death was 82 days. For the Delphi process, 12 oncology or palliative care health professionals were included. A consensus was reached on the minimum data to be recorded in case of PS. This list includes: selection criteria, decision-making process and the sedation evolution. CONCLUSIONS: PS was applied in half of the patients who died due to dyspnea or delirium. Selection criteria were identified, as well as the type of PS and patient involvement in decision making process. A consensus was also reached on a minimum dataset that would help the clinician to record relevant information in PS.