Code status options, how they are named and defined, vary depending on the hospital, according to the results of a recent study.1 “Hospitals are doing this differently from one another. And we have no idea what the consequence is,” says lead study author Jason Neil Batten, MD, a researcher at the Stanford Center for Biomedical Ethics.

As just one example, each of the seven hospitals in the study named and defined DNR orders somewhat differently. “We have decades of research showing that people do not always understand the very same DNR order in the same way. Now, we find that a DNR at one hospital is very different from another hospital,” Batten explains.

Stanford Hospital had changed the code status system and policies about a decade ago at the adult hospital, and made a similar change at the children’s hospital several years later.

“But we had no data about whether the changes were good, or even what other hospitals were doing, other than a few anecdotes,” says David Magnus, PhD, the study’s senior author and director of the Stanford Center for Biomedical Ethics.

For example, a “Do Not Escalate” (DNE) order was added. The change was popular with some clinicians but opposed by others.

“We decided that a qualitative study with purposive sampling would be a good way to proceed,” Magnus says. There was much greater variation than expected across the different institutions. Code status ordering systems were asked to perform different functions. Some specified what interventions will be limited. Some specified whether a patient will be sent to the ICU or a higher level of care. Others specified an overall philosophy of care.

“The results suggested that the code status ordering systems can sometimes contribute to goal-discordant care and miscommunication about patient wishes,” Magnus reports.

One or more clinicians speaks to the patient or surrogate, then translates the goals of care into a set of orders. Later, other clinicians interpret those orders to make concrete decisions. “That process is fraught with potential for misalignment,” Magnus says. “Our findings suggest that there are better and worse systems for avoiding these problems.”

The researchers followed up by testing Stanford’s own code status system to see whether physicians would put in the same orders for the same scenarios or take the same actions for the same scenario and code status. “We found significant rates of discordance,” Magnus observes.

Whether a patient was intubated or sent to the ICU varied depending on the physician. Based on these data, Stanford designed and tested a new code status system. It has dramatically reduced the degree of discordance. “We hope that other hospitals will consider our findings and do more actual testing of how well their systems really work at achieving their goals,” Magnus offers.

Currently, there really is no hard evidence indicating which code status options are best. “I have my opinions as a clinician, but I can’t tell you as a researcher which way is better, or even the effects of it,” Batten admits. “None of this has been adequately researched. We just don’t know.”

Traditionally, a DNR order is supposed to be narrowly focused on what to do if the patient is in cardiac arrest. “It’s not supposed to have any impact on any other areas of their care,” Batten says.

But at Stanford, the “DNR/DNE” option carries broader implications. It means the overall philosophy of care is comfort care.

“At a place like Stanford, the code status options have a very different role. They are not just telling you what to do for cardiac arrest. In some cases, they are guiding the entire trajectory of a patient’s care,” Batten says.

Some ethics consults revealed unintended consequences of the new code status options. For instance, nurses questioned if they should give pain medications to a patient who is DNR/DNE, who is not at the point of comfort care, but with symptoms not as well-controlled as they should be. Clinicians also wondered if treating pneumonia is compatible with comfort care, or if changing antibiotics counted as “escalation.”

“There are also some pretty straightforward patient safety issues in accurately capturing the code status orders and how those were interpreted in an emergency scenario,” Batten adds.

This can happen if a patient wants to be intubated for respiratory distress, but is not intubated because different people understood a Do Not Intubate (DNI) order in different ways. “If you as a patient thought that DNI meant something, and the clinician thinks it means something different, that means we thought we knew your code status but you still didn’t get what you wanted,” Batten explains.

When patients move between hospitals, variations present other challenges. Patients might come from one hospital with a narrowly defined DNR order, then they come to Stanford where there are several different DNR orders. Clinicians have to “translate” which one is appropriate for that person. A well-documented discussion on goals of care can help in that regard. Ideally, the patient is not just handed a list of code status options to pick and choose from. “In some places, this is just becoming like a menu that’s guiding the conversation,” Batten says.

Patients really do not need to know all the different terminology of code status options. A clinician might put it this way: “If your heart were to stop, we will just let you pass peacefully and we’re not going to add any more intensive therapies to your care.”

“That’s a patient-friendly way to say ‘You are a DNR/Do Not Escalate,’” Batten notes.

There can be confusion even among clinicians. Some medical trainees learn the code status options at Stanford, then go practice at a different hospital. Others practice at multiple institutions. “A lot of them are not that aware that code status functions differently at different hospitals,” Batten says.

The same is true of clinical ethicists. Some ethicists are unaware of how code status options are named and defined at their institutions. Notably, the researchers found that at some hospitals, code status options in the electronic medical record did not even match what was in the hospital policy. “Ethics should be one of the experts on this — maybe not the only expert, but you should be able to ask the ethics committee for the policy, and what are the options in the electronic medical record, and are they the same?” Batten argues.

Ethicists also can consider the scope of DNR orders, whether they are narrowly focused on cardiac arrest, or if they broadly address all aspects of care — and if it is clear to everybody. “You really need to do QI work at your institution, to get out there and understand how these things are functioning,” Batten says. “That can improve the quality of care on a systems-level basis.”


  1. Batten JN, Blythe JA, Wieten S. Variation in the design of Do Not Resuscitate orders and other code status options: A multi-institutional qualitative study. BMJ Qual Saf 2020 Oct 20;bmjqs-2020-011222. doi: 10.1136/bmjqs-2020-011222. [Online ahead of print].