Retrospective studies of end-of-life resource utilization and costs in cancer care using health administrative data: a systematic review

Background: There has been an increase in observational studies using health administrative data to examine the nature, quality, and costs of care at life's end, particularly in cancer care.

Aim: To synthesize retrospective observational studies on resource utilization and/or costs at the end of life in cancer patients. We also examine the methods and outcomes of studies assessing the quality of end-of-life care.

Design: A systematic review according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and AMSTAR (A Measurement Tool to Assess Systematic Reviews) methodology.

Data sources: We searched MEDLINE, Embase, CINAHL, and York Centre for Research and Dissemination (1990-2011). Independent reviewers screened abstracts of 14,424 articles, and 835 full-text manuscripts were further reviewed. Inclusion criteria were English-language; at least one resource utilization or cost outcome in adult cancer decedents with solid tumors; outcomes derived from health administrative data; and an exclusive end-of-life focus.

Results: We reviewed 78 studies examining end-of-life care in over 3.7 million cancer decedents; 33 were published since 2008. We observed exponential increases in service use and costs as death approached; hospital services being the main cost driver. Palliative services were relatively underutilized and associated with lower expenditures than hospital-based care. The 15 studies using quality indicators demonstrated that up to 38% of patients receive chemotherapy or life-sustaining treatments in the last month of life and up to 66% do not receive hospice/palliative services.

Conclusion: Observational studies using health administrative data have the potential to drive evidence-based palliative care practice and policy. Further development of quality care markers will enhance benchmarking activities across health care jurisdictions, providers, and patient populations.