The documents at issue are Physician/Portable Orders for Life-Sustaining Treatment, known by the acronym POLST and by related names across the country. POLST was initially developed in 1991 after clinicians and ethicists recognized that the treatment preferences of seriously ill patients were not consistently available nor honored during transitions and emergencies. Over the following decades, jurisdictions developed individual forms under multiple titles, including Medical Orders for Life-Sustaining Treatment (MOLST), Portable Orders for Sustaining Treatment (POST), and Medical Orders for Sustaining Treatment (MOST). In 2025, the National POLST Collaborative revised the title to emphasize POLST as portable medical orders, reflecting a scope that extends beyond CPR to other treatment decisions and goals of care. Completion of a POLST is voluntary, and healthy adults are not eligible. A completed form documents treatment choices that have been discussed and translated into medical orders, with signatures of the patient or their spokesperson and the provider indicating agreement with the selected options. This distinguishes POLST from an advance directive, which is a broader legal document available to adults who wish to name a proxy and state future health care preferences; POLST converts treatment preferences for persons with serious illness or advanced frailty into orders that clinicians can implement.

CPR provides a useful analytic case for examining these forms as communication tools because the decision is time-sensitive and is usually encountered early on the form. In almost all jurisdictions, the first category of orders is a directed response to cardiopulmonary resuscitation. POLST forms are single-page documents that can be printed on an 8.5 by 11 inch page, and within this limited space the CPR decision appears alongside patient-identification information, instructions, and orders addressing other aspects of care. On the form created by the National POLST Collaborative, the CPR decision occupies approximately 14 percent of the upper visual field. Across other jurisdictional forms, both the amount of space devoted to the CPR decision box and the amount of information surrounding it vary. The authors reviewed the forms as communication documents rather than legal instruments, and the analysis is explicitly descriptive rather than evaluative: forms are not ranked and no layout is assumed to be optimal.

To organize the evaluation, the team defined a Find–Understand–Act framework, a newly defined variation on generic logic model development. Find asks whether a responder can readily locate the CPR decision within the surrounding information. For each form, researchers measured usable page width, upper-field height, and CPR-field height in inches, using electronic copies of the printable forms and Adobe measurement tools. The upper field extended from the first printed content on page one through the bottom of the CPR decision field, and derived measures included upper-field area, CPR-field area, and the proportion of the upper-field area occupied by the CPR field. Word counts were recorded separately for the upper field and CPR field, and word density was calculated as words per square inch for each area. Selected forms were measured by two individuals to assess consistency, and discrepant measurements or word counts were adjudicated during review by the full team. The team elected not to capture font size, text spacing, and other aspects of typography because variation across jurisdictions did not lend itself to a systematic measurement approach.

A key conceptual contribution of the Find stage is the separation of two visual characteristics that are often conflated: prominence and textual density. A scatterplot of word density within the CPR box against the CPR field as a portion of the entire upper field showed that these measures were not interchangeable. Forms could allocate substantial space to CPR yet contain relatively little text, or provide less space while presenting greater textual burden. Three extreme forms were omitted from the publication-facing scatterplot to preserve visibility of the central distribution but were retained in descriptive analyses and discussed anonymously. Jurisdiction names were retained in the analytic dataset for verification and adjudication but were not used in publication-facing figures, and Oregon was excluded from replicated publication figures because of restrictions on reproduction of the form. This distinction is particularly relevant to portability, because responders may encounter forms from jurisdictions other than their own, and the National POLST Collaborative notes that local laws, regulations, and approved forms differ even when programs share a common conceptual model.

Understand begins after the decision box has been found and asks whether the responder can identify the physiologic condition that activates the order and correctly associate that condition with the authorized treatment choice. The researchers recorded the clinical trigger used to define when the CPR order applies, reviewing trigger language for explicit reference to absence of pulse, absence of breathing, cardiopulmonary arrest, or other physiologic conditions. Their findings show that most forms converged on a common clinical state, anchoring the CPR decision to a shared physiologic condition, but that trigger wording was not linguistically uniform. There was considerable variation in terminology, labels, and the degree to which interpretation depended on cross-references, adjacent sections, or additional descriptors. Notably, the authors treated terms such as unresponsive as not equivalent to pulseless or not breathing, because unresponsiveness alone cannot establish cardiopulmonary arrest. The communication task therefore extends beyond locating a prominent yes/no choice to interpreting the relationship between the clinical trigger and the treatment instruction.

Act extends the framework from interpretation to implementation, examining whether and how the CPR field communicates the response expected after the selected order is identified and understood. The researchers extracted responder-directed language attached to affirmative and negative CPR choices, including formulations such as Attempt Resuscitation, Attempt CPR, Do Not Attempt Resuscitation, Do Not Resuscitate, No CPR, and Allow Natural Death. A frequency-weighted phrase display of actions associated with each choice revealed a marked asymmetry. A yes-CPR response directed initiation of resuscitative actions, and affirmative phrases commonly named related treatment actions. A no-CPR response, by contrast, was dominated by instructions not to initiate resuscitation, with few words describing clinical care after the decision was implemented. The authors call this asymmetry an action-language gap, a communication boundary requiring further study that also presents a challenge to the concept of portable medical orders. The instruction Follow Orders was retained as a bridging phrase in the display because it could support either CPR choice by directing the responder to act in accordance with the documented medical order.

The action-language gap matters because No CPR does not mean no care. The physiologic trigger specifying a no-resuscitation order represents the end of a process: the order applies when the heart has stopped and the person is not breathing, and the decision is made prior to the event. Once a decision is made not to resuscitate, other clinical needs remain and are part of the dying process. Active dying may include irregular breathing, mottling, deep unresponsiveness, low urine output, and death rattle. Comfort-focused care represents an affirmative clinical response to these symptoms rather than an absence of intervention, and the authors argue that the Act language for no-CPR orders needs to communicate care targeting the comfort needs of persons at the end of life who have elected not to receive resuscitation. The gap hints at health literacy issues for the comfort care envisioned by both health care professionals and people selecting no CPR, and the authors note that POLST may deliver its greatest benefit upstream of the acute care setting, including prehospital settings and nursing homes, where it can interrupt clinical momentum toward unwanted treatment and help align care with documented wishes.

The authors are careful to state the limits of what this analysis establishes. It is primarily a descriptive communication analysis rather than an assessment of legal validity, clinical effectiveness, responder performance, or patient outcomes. Measurements describe features of the forms but do not demonstrate that one design is easier to use or produces better decisions in emergencies, and observed differences should not be interpreted as measures of form quality. The analysis was restricted to publicly available printable POLST forms; electronic or digitally rendered formats, registry interfaces, and electronic health record displays were not evaluated. The analytic unit was the portion of the page containing the CPR decision and the material appearing above and immediately adjacent to that field, with broader treatment-plan sections examined only when explicitly linked to the CPR decision. Simulation or field-based research would be required to determine how the observed differences affect responder recognition, interpretation, and action under time pressure. What the study does establish is a common measurement framework applied across jurisdictional forms, pairing quantitative visual measures with examination of clinical trigger language, treatment-choice wording, and responder-directed action, and it identifies a concrete communication boundary that warrants further study in the design of portable medical orders.

Subject of Research: A communication analysis of CPR sections on POLST forms across 51 U.S. jurisdictions using a Find-Understand-Act framework

Article Title: Finding, understanding, and acting on CPR orders: a communication analysis of POLST forms across 51 jurisdictions

Article References: Miles, T. P., Catron, S., Brown, M., Lavender, A., & Bayakly, R. (2026). Finding, understanding, and acting on CPR orders: a communication analysis of POLST forms across 51 jurisdictions. Frontiers in Public Health, 14, Article 1979970. https://doi.org/10.3389/fpubh.2026.1979970

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1979970

Keywords: POLST, CPR, portable medical orders, end-of-life care, health communication, resuscitation, advance care planning, health literacy, palliative care, medical orders, Finding, understanding