Preference for place-of-death among terminally ill cancer patients in Denmark

Authors


Mette Asbjoern Neergaard, The Palliative Team, Aarhus University Hospital, Noerrebrogade 44, 8000 Aarhus C, Denmark.
E-mail: man@alm.au.dk

Abstract

Scand J Caring Sci; 2011; 25; 627–636

Preference for place-of-death among terminally ill cancer patients in Denmark

Achieving home death is often seen as an important endpoint in palliative care, but no studies of the preferred place-of-death have yet been conducted in Scandinavia. Furthermore, we do not know if professionals’ report on deceased patients’ preference of place-of-death is a valid information. The aim of this study was to describe where terminally ill Danish cancer patients prefer to die and to determine if their preference changed during the palliative period, as reported retrospectively by bereaved relatives, general practitioners (GPs) and community nurses (CNs) and to assess the agreement of their accounts. The study was a population-based, cross-sectional combined register and questionnaire study in Aarhus County, Denmark. The population comprised 599 deceased adult cancer patients who had died from 1 March to 30 November 2006 and were identified through merging of health registers. Relatives returned 198 questionnaires about patients’ preferred place-of-death, GPs 333 and CNs 201. The study showed that most terminally ill cancer patients preferred home death (up to 80.7%). The reported preference for home death weakened as death approached (down to 64.4%). A better congruence was seen between relatives’ and GPs’ accounts of preference for place of death at the end of the palliative period (κ 0.71) than between relatives’ and CNs’ accounts (κ 0.37). In conclusion, bereaved relatives (and GPs and CNs) report retrospectively that most terminally ill cancer patients wish to die at home. The preference weakened significantly as death approached. The agreement between relatives’ and GPs’ accounts on patients’ preferences at the end of the palliative period was ‘substantial’, whereas the agreement between relatives’ and CNs’ accounts at the same time was significantly less outspoken. This indicates that CNs may be facing a problem in assessing their patients’ wishes retrospectively.

Ancillary