Presentation (CTI)

OBJECTIVES: To describe healthcare utilization trajectories from life-sustaining treatment (LST) documentation to death and to compare end-of-life care settings according to implementation of LST decisions.
METHODS: We conducted a retrospective cohort study using Korean National Health Insurance claims data. Patients with documented LST plans who were followed until death were included. Healthcare utilization from LST documentation to death was reconstructed on a patient-day basis. Care states were classified as tertiary hospital inpatient care, general hospital inpatient care, other hospital inpatient care, long-term care hospital, hospice ward, hospice dying room, home hospice, and outpatient/other. Baseline characteristics were compared using Wilcoxon rank-sum tests for continuous variables and chi-square tests for categorical variables. Landmark state distributions at 30, 14, 7, 3, and 1 days before death and the last transition before death were summarized.
RESULTS: The cohort included 408,721 patients; 372,200 (91.1%) were in the implementation group and 36,521 (8.9%) in the non-implementation group. Median follow-up from LST documentation to death was 20 days (IQR 7-51) in the implementation group and 44 days (IQR 19-118) in the non-implementation group. Mean hospice ward occupancy was 0.157% versus 0.293%, and mean home hospice occupancy was 0.208% versus 0.928%, respectively. At 30 days before death, outpatient/other accounted for 94.5% of patient-days in the implementation group and 94.3% in the non-implementation group. Home hospice accounted for 0.4% and 1.1%, and hospice ward for 0.2% and 0.3%, respectively. The most frequent last transition in both groups was outpatient/other to death.
CONCLUSIONS: Across the period from LST documentation to death, non-hospice states accounted for most patient-days, while hospice-related occupancy remained low overall and was lower in the implementation group than in the non-implementation group.