Regional differences in do-not-resuscitate orders track with how Americans spend their final days
Regional practice patterns still shape the way death unfolds in the United States, and do-not-resuscitate orders remain one of the clearest markers of that variation. This European Heart Journal abstract focuses on the relationship between regional differences in DNR orders and end-of-life health care use and spending, a question with direct relevance for clinicians, families, and health systems trying to align care with patient goals.
The premise is clinically familiar. A DNR order does not represent abandonment; it defines the boundary of attempted resuscitation when cardiopulmonary arrest occurs. Yet in real-world practice, the presence or absence of that order often sits inside a broader pattern of decision-making about intensive care, hospital utilization, and the intensity of treatment delivered near the end of life. In other words, a single line in the chart can reflect a much wider conversation about prognosis, preferences, and the expected burdens of intervention.
Why regional DNR patterns matter beyond the code status itself
Differences in DNR use can arise from patient values, clinician communication, local norms, hospital culture, and access to palliative care. When those forces vary by geography, the downstream effects may show up in measurable ways: emergency department visits, ICU admissions, in-hospital deaths, and spending during the final phase of life. End-of-life care is one of the few settings in medicine where resource use, symptom burden, and decisional quality are tightly intertwined.
A region with lower DNR use may not simply be delivering “more care.” It may be sustaining more aggressive treatment late into a trajectory where recovery is no longer realistic, especially if conversations about goals of care are delayed until the patient is already critically ill. A region with higher DNR use may still provide appropriate, symptom-focused treatment, but with less exposure to invasive resuscitation and the cascade of interventions that often follows cardiac arrest or escalating organ failure.
The question is not whether DNR orders save money in isolation. The more important issue is whether they function as a signal of earlier, clearer decision-making that helps avoid unwanted treatment while preserving comfort and dignity. Spending becomes a downstream metric of those choices, not the moral center of them.
End-of-life utilization often reflects how late decisions are made
Near the end of life, the timing of goals-of-care discussions can be as important as the decisions themselves. When these conversations happen early, patients may receive treatments that match their preferences, including hospice referral, symptom control, and less time in high-intensity settings. When they happen late or not at all, clinicians often default to life-prolonging interventions because there is no documented alternative path.
That dynamic helps explain why regional differences in DNR orders can matter for health care spending. Intensive care, repeated hospitalizations, procedures, and prolonged inpatient stays are expensive, but they also carry a high emotional and physical toll when the likelihood of meaningful recovery is low. In the final weeks or months of life, the clinical question is rarely whether medicine can do more. It is whether doing more still helps the person in front of the team.
The abstract’s emphasis on U.S. regional variation suggests that the study is examining more than individual preference. It points toward structural differences in how care is delivered and documented. Such variation can reveal inequities in communication, in access to palliative services, and in the ability of patients to have their wishes translated into action before a crisis occurs.
Spending is the visible footprint of a larger clinical process
Health care spending at the end of life is often treated as a policy problem, but the medical meaning is more immediate. High spending can reflect repeated transitions, fragmented care, and the default use of hospital-based rescue treatment even when benefit is doubtful. Lower spending does not automatically mean better care, but in the context of well-documented goals and appropriate symptom management, it can indicate a more coherent alignment between treatment intensity and patient priorities.
For cardiology in particular, this topic carries added weight. Cardiac arrest, advanced heart failure, and terminal arrhythmias are conditions in which resuscitation decisions are immediate and consequential. A DNR order does not change the underlying disease, but it changes the response when the disease reaches a terminal point. It also shapes how the final interaction is experienced by the patient and by the team at the bedside.
The abstract points to a systems problem, not a paperwork issue
Code status documentation is sometimes treated as administrative formality. It is not. It is a clinical expression of what kinds of intervention a patient wants when their body can no longer sustain itself. If regional differences in DNR orders are associated with different patterns of end-of-life use and spending, then the finding speaks to how effectively the health system converts preferences into practice.
That has practical implications for hospitals, physicians, and policy leaders. Better advance care planning, earlier serious-illness conversations, and more reliable palliative care integration may reduce unwanted treatment and improve the consistency of care delivery across regions. The aim is not to suppress treatment. The aim is to make sure treatment remains proportionate to prognosis and consistent with patient wishes.
A narrow abstract cannot answer every question. It does not reveal which regions were compared, how DNR use was measured, or how much of the spending difference was attributable to hospitalization, ICU care, or other services. It also cannot establish causality from regional association alone. Still, the clinical message is unmistakable: end-of-life care in the United States remains geographically uneven, and DNR orders appear to be one visible marker of that unevenness.
The next step is less about asking whether variation exists and more about asking why it persists in a system that already knows how to document patient preferences. Until that gap closes, the final chapter of care will continue to depend too heavily on where a patient happens to be treated.