• Katrin R Sigurdardottir, Stein Kaasa, Jan H Rosland, Claudia Bausewein, Lukas Radbruch, Dagny F Haugen, and PRISMA.
• 1 European Palliative Care Research Centre, Department of Cancer Research and Molecular Medicine, Faculty of Medicine, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.
• Palliat Med. 2014 Jun 1; 28 (6): 463-473.

BackgroundOne of the barriers identified in palliative care research is the lack of common criteria to describe the population.AimThe aim of this Delphi process was to obtain consensus on a basic set of core variables to describe or classify a palliative care cancer population.Design And SettingThis was a five-step international Delphi exercise. A total of 117 experts were invited to participate. Based on a literature review and analyses of existing minimum datasets for national databases, a list of 18 proposed variables was presented in the first Delphi round. The two first rounds focused on which variables to include, and several new variables were proposed. The three last Delphi rounds focused on how the agreed variables should be recorded. Consensus was defined as at least 70% agreement.ResultsA total of 64 experts from 30 countries participated. High consensus was reached on 31 variables, divided between a 'patient form' - date of birth, gender, living situation, education, ethnicity and 12 symptoms - and a 'health-care personnel form' - patient's date of birth, principal diagnosis, date of the principal diagnosis, stage of the cancer disease, site of metastases, present anticancer treatment, main additional diagnoses, stage of the additional diagnoses, medication, weight loss, performance status, cognitive impairment, place of care and provision of care. It was more difficult to agree upon how to record the variables, but consensus was reached on all except ethnicity, vomiting and weight loss.ConclusionConsensus was reached on a set of core variables and how they should be recorded.

5 keywords...

hide…