Place of care in advanced cancer: a qualitative systematic literature review of patient preferences.

BACKGROUND: It is commonly written that more patients wish to die at home than currently achieve this. However, the evidence for preferences for place of terminal care and death has not been systematically reviewed. AIM: To carry out a systematic literature review of the preferences for place of care and death among advanced cancer patients.
METHOD: Studies were identified using systematic database searches of MEDLINE (1966-1999), PsychLit (1974-1999), and Bath Information Data Service (BIDS) (1981-1999). Studies were assessed and data extracted and synthesises following the NHS Centre for Reviews and Dissemination guidelines, grading studies according to design and rigor of methods. Studies of preferences in the general population and of groups including cancer patients and/or their caregivers were included.
RESULTS: Eighteen studies determining preferences in either the general population or groups including cancer patients were identified. Views were obtained prospectively and retrospectively from patients, the general population, families, and professionals. Respondents indicated preferences for home death (range 49%-100%), except one study of patients in the care of a continuing care team in London where only 25%-29% of patients wanted a home death, and inpatient hospice was the most favored option. However, the response rate of this study was not known. Among the general public there was a higher preference for inpatient hospice care among in people with recent experience of a close friend or relative's death or dying. Where the views of patients, families, and professionals were compared, all respondents broadly agreed although patients expressed the strongest home preferences. Only 2 of the studies provided longitudinal data, and 9 of the 18 had major deficits in design or reporting, such as poor or unknown response rate, unclear or unsystematic methods of eliciting preferences or other sample or measurement bias. However, sensitivity analysis of only the more robust and larger studies did not alter the finding of a preference for home care at the end of life in over 50% of patients.
CONCLUSIONS: Home care is the most common preference, with inpatient hospice care as second preference in advanced illness. Meeting these preferences could be important outcomes for services. Study designs in this area need to be improved.