• Jane Gibbins, Colette M Reid, Sophia Bloor, Melanie Burcombe, Rachel McCoubrie, and Karen Forbes.
• Cornwall Hospice Care, St. Julia's Hospice, Hayle, Cornwall. janegibbins@hotmail.com
• J Pain Symptom Manage. 2013 May 1;45(5):859-67.

ContextApproximately 40% of the U.S. and 60% of the U.K. population die in hospital. Many reports have highlighted variability in the care received by these patients and national initiatives have proposed strategies to improve this care. No studies have demonstrated whether any improvements in end-of-life care have been achieved, as research in this area is challenging.ObjectivesWe designed a study to assess the feasibility of a novel method of identifying patients likely to die during an acute hospital admission and a model of prior consent from patients and/or assent from their relatives.MethodsA study for collecting data on patients' symptoms before and after the introduction of an end-of-life tool (comprising medical and nursing checklists, prescribing guidance, and a symptom observation chart) within five wards in a major U.K. teaching hospital was conducted. We asked the screening question to a senior member of staff, "Is this patient so unwell that you feel they could die on this admission?" to identify appropriate patients, and recruited using the consent procedure. Patients were enrolled in the study if they became more unwell and data were then collected until they died.ResultsIn total, 6642 patients were screened. The ward staff answered "yes" to the screening question for 327 of 6642 (4.9%) patients. Patient's prior consent or relative's assent to enroll in the study was obtained for 117 of 327 (35.8%) patients, of whom 70 of 117 (59.8%) enrolled for the study and died within the study period. The staff found that the methods used were appropriate.ConclusionWe have shown that identifying and involving dying patients in research is possible and acceptable to patients and carers.Copyright © 2013 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.

28 keywords...

hide…