Objectives: To evaluate whether terminal care was consistent with Physician Orders for Life-Sustaining Treatment (POLST), a preprinted and signed doctor's order specifying treatment instructions in the event of serious illness for CPR, levels of medical intervention, antibiotics, IV fluids, and feeding tubes.
Design: Retrospective chart review.
Setting: ElderPlace, a Program of All-Inclusive Care for the Elderly (PACE) site in Portland, Oregon.
Participants: All ElderPlace participants who died in 1997 were eligible (n = 58). Reasons for exclusion were no POLST (1), missing POLST (1), and insufficient documentation of care (2).
Measurements: POLST instructions for each participant and whether or not each of the treatments addressed by the POLST was administered in the final 2 weeks of life.
Results: The POLST specified "do not resuscitate" for 50 participants (93%); CPR use was consistent with these instructions for 49 participants (91%). "Comfort care" was the designated level of medical intervention in 13 cases, "limited interventions" in 18, "advanced interventions" in 18, and "full interventions" in 5. Interventions administered were at the level specified in 25 cases (46%); at a less invasive level in 18 (33%), and at a more invasive level in 11 (20%). Antibiotic administration was consistent with POLST instructions for 86% of 28 subjects who had infections in the last 2 weeks of life, and less invasive for 14%. Care matched POLST instructions in 84% of cases for IV fluids and 94% for feeding tubes.
Conclusions: POLST completion in ElderPlace exceeds reported advance directive rates. Care matched POLST instructions for CPR, antibiotics, IV fluids, and feeding tubes more consistently than previously reported for advance directive instructions. Medical intervention level was consistent with POLST instructions for less than half the participants, however. We conclude that the POLST is effective for limiting the use of some life-sustaining interventions, but that the factors that lead physicians to deviate from patients' stated preferences merit further investigation.