Do the preferences of patients with terminal cancer or their family caregivers and end-of-life care discussions influence utilization of hospice-palliative care?

Abstract

9041 Background: We aimed to determine whether utilization of hospice-palliative care (HPC) was associated with end-of-life (EOL) care discussions and the stated preferences of patients and their family caregivers, a topic dealt with in few studies. Methods: As part of the Study to Understand Risks, Priority, and Issues at End-of-Life (SURPRISE), a prospective multicenter cohort study, we administered similar questionnaires to 619 patients and their family caregivers at the time of the terminal diagnosis and a follow-up questionnaire to 363 caregivers 3 months after the patient died. The baseline questionnaire covered demographic characteristics, religion, EOL care preference (HPC, intensive care, and cardiopulmonary resuscitation [CPR]), and EOL care discussions held with physicians concerning outpatients or inpatient visits, utilization of HPC, home care services, the intensive care unit (ICU), CPR, or social services. The follow-up questionnaire determined actual utilization of HPC. Results: A total of 257 (70.8%) family caregivers completed the second questionnaire and 104 (40.5%) of their patients utilized HPC. The multivariate model showed that the family caregivers’ preference for HPC was associated with its actual utilization (adjusted odds ratio [aOR], 2.07; 95% confidential interval [CI], 1.14 to 3.75), whereas the patients’ preference was not. Patients were more likely to utilize HPC if family caregivers had discussed with physicians either HPC (aOR, 2.34; 95% CI, 1.29 to 4.24) or ICU admission (aOR, 2.41; 95% CI, 1.05 to 5.55). Another factor associated with HPC utilization was the religion of the family caregiver (Catholic or Protestant vs. Buddhist or none, aOR, 1.97; 95% CI, 1.13 to 3.44). No patient or demographic-clinical factor was related to HPC utilization. Conclusions: HPC utilization appears to be influenced by family caregiver (rather than patient) preferences and by EOL care discussions between physicians and family caregivers.