Background: Portable Orders for Life-Sustaining Treatment (POLST) forms are an integral patient-centered care tool which facilitate conversation about and documentation of patients’ preferences for life-sustaining treatment. As the proportion of hospitalized patients with serious progressive illness increases over time, advanced care planning has become paramount. Within the adult acute care population, however, POLST form completion occurs infrequently and thus may increase the risk of undesired resuscitation and intensive care.1-3 The aim of this quality improvement project is to identify opportunities for and improve rates of POLST form completion among hospitalized adults on a general medicine service at a single academic tertiary referral hospital.
Methods: Using local POLST registry data we calculated the baseline prevalence of adults aged 50 or older with a do-not-resuscitate (DNR) order and concordant POLST form at time of hospital discharge by an internal medicine physician. Patients with full code status and those with pre-existing POLST forms with DNR orders were excluded. PDSA cycles to increase in-hospital POLST completion included a March 2020 training session for internal medicine hospitalists and an EHR discharge navigator notification that went live in June 2020. Data were tracked through September 2020 via a control chart. Binomial proportion tests were performed to determine significant differences after interventions using RStudio (version 1.2.5001; 2009-2019 RStudio, Inc).
Results: Within the pre-intervention and intervention period from July 2019 to September 2020, there were 178 patients with DNR code status but without a concordant POLST form who were hospitalized at this institution. Pre-intervention POLST form completion rate for the target patient population was 28.8% based on data from July 2019 to March 2020. The first intervention and PDSA cycle (April-June 2020) did not significantly change completion rates (20.0% from baseline 28.8%, p=0.41). POLST completion rates re-approached baseline rates after the second intervention and PDSA cycle (July-September 2020) but did not reach statistical significance (27.8% vs 28.8%, p=0.99).
Conclusions: This quality improvement initiative demonstrated a baseline POLST completion rate for adult internal medicine patients aged 50 years and older hospitalized with DNR-at-discharge status of ~29%, which is consistent with available data in similar settings. Group education and an electronic health record (EHR) alert were ineffective short-term interventions to improve completion rates. We suspect that EHR notification fatigue and lack of individualized performance feedback were explanatory. Additional PDSA cycles continue and include quarterly email notifications to internal medicine clinicians that share their individual performance in relation to anonymized peers.