Elderly Patients Presenting to a Level I Trauma Center with a POLST form: A Propensity-Matched Analysis.

BACKGROUND Physician Orders for Life-Sustaining Treatment (POLST) forms are portable medical orders documenting patient treatment preferences in an acute health decline. It is unclear how these forms are used in the management of elderly trauma patients. METHODS Patients 65 years and older presenting to a Level I trauma center were identified between 2012 and 2017. Hospital trauma registry and medical records were used to identify a preinjury POLST and its acknowledgment by providers within 24 hours of arrival. A 1:1 propensity score matched sample was used to evaluate clinical outcomes based on the presence of a POLST limiting interventions with p less than 0.05 deemed significant. RESULTS There were 3,342 elderly trauma patients identified. One hundred ninety-two (6%) had a POLST identified by the institutional trauma registry dated before the injury. Do not attempt resuscitation (DNR) was listed in 154 patients (80%), and 79% desired to avoid the intensive care unit (ICU) with limited (54%) or comfort measures only (CMO, 25%). One hundred seven (76%) of admitted POLST DNR patients had a DNR code status for the majority of their admission. 59 (58%) of the limited and 29 (60%) of the comfort measures only patients were admitted to the ICU. Acknowledgment of a preinjury POLST or code status was explicitly documented in 110 cases (57%). Propensity score analysis yielded a comparison sample of 288 patients. In the matched comparison, an acknowledged POLST with limitations was associated with a shorter ICU stay (1.7 vs. 2.8 days, p = 0.008) but there was no difference in ICU admission (58% vs. 61%, p = 0.69), total length of stay (3.8 days vs. 4.8 days, p = 0.08), or in-hospital mortality (13% vs. 8%, p = 0.2). CONCLUSION Limited provider acknowledgment of preinjury medical directives necessitates protocol development for the management of frail elderly trauma patients. When acknowledged, patients with a POLST limiting interventions had fewer ICU days without increased in-hospital mortality compared with similarly injured elderly patients. Copyright (c) 2019 Wolters Kluwer Health, Inc. All rights reserved.