Goal-Concordant Care Among Seriously Ill Hospitalized Patients With Treatment Limitations on POLST: A Mixed-Methods Study

Abstract

Background: POLST enables patients with serious illness to document care preferences. POLST-discordant care is common, but little is known about the context in which this occurs and whether it represents goal-discordant care. We sought to examine the context in which seriously ill patients receive POLST-discordant care and inform efforts to align end-of-life care with patients' preferences. Methods: We performed a chart review of the electronic medical records of patients with POLST preferences for limited treatment or comfort measures who presented to the emergency department and received high-intensity treatment (i.e., POLST-discordant care) between April 2015 and October 2016 at a quaternary medical center. High-intensity treatment was defined using a validated set of measures. We used inductive/deductive content analysis to analyze documentation pertaining to patients' hospitalizations and identify themes related to medical decision making and goals-of-care discussions. Results: A total of 32 patients with treatment limitations on POLST received POLST-discordant care. Cancer, cerebrovascular disease, and heart failure were the most prevalent comorbidities. Of these, 30 patients (94%) presented with acute surgical diagnoses and underwent ≥ 1 procedure; 24 (75%) underwent intubation/mechanical ventilation, and 14 (44%) were admitted to the ICU. We identified four themes related to the receipt of POLST-discordant care: (1) POLST-discordant care may represent goal-concordant care; (2) “Fix It” mindset among clinicians contributes to high-intensity treatment; (3) Incomplete understanding of clinical trajectory may contribute to non-beneficial care; and (4) Challenges in POLST implementation limit its utility to address context-specific medical decision-making. Conclusions: This mixed-methods study of seriously ill patients with treatment limitations on POLST provides context for receipt of “POLST-discordant care” in the acute-care setting and reveals how such treatment may still represent goal-concordant care. Our findings reveal complexities when considering goal-concordant but potentially non-beneficial care among seriously ill patients, and demonstrate the impact of clinicians' mindsets and communication on delivery of high-intensity treatment. Finally, our study highlights the nuances of medical decision-making and the persistent challenge of interpreting advance-care planning documents in acute-care settings.