Case and Commentary
Oct 2012

Why Not a Slow Code?

Edwin N. Forman, MD and Rosalind E. Ladd, PhD
Virtual Mentor. 2012;14(10):759-762. doi: 10.1001/virtualmentor.2012.14.10.ecas1-1210.


Dr. Holbert, a senior neonatologist, was called to attend a patient he knew very well, Mrs. Gage, during the birth of her first child. She was at the hospital with her husband and about to deliver the child, who had been prenatally diagnosed with a number of severe deformities, including a congenital diaphragmatic hernia and severe cardiac anomalies. Dr. Holbert had had a number of frank conversations with Mrs. Gage and her husband and informed them that their child’s anomalies were very severe and possibly life threatening. Despite this, she and her husband uniformly insisted that a full and aggressive resuscitation be undertaken. At the end of their most recent discussion, Mrs. Gage agreed to let the team make their initial assessment at the time of delivery and discuss with her and her husband how to proceed at that time. Dr. Holbert felt this was a reasonable plan.

With his resuscitation team in place, Dr. Holbert received the newborn baby boy, who had a weak pulse and was making labored respiratory efforts. Upon reaching the infant warmer to assess the child, Dr. Holbert realized that the anomalies were more severe than initially expected, and, though they might be able to keep him alive temporarily on maximal support, he would most likely never leave the ICU. When Dr. Holbert reported to Mr. and Mrs. Gage on their son’s condition and his prognosis, they said, “We want you to do everything. Please don’t let our son die.”

Trying to balance the good of the child and the emotional needs of the parents, Dr. Holbert turned to his team and quietly instructed them to undertake a “slow code.”


Many medical students, residents, and other medical staff learn the elements of a slow code early in their clinical years. It seems to be part of the wisdom that some experienced physicians pass on, following a long history known among the medical profession but not generally known to the public.

The intentions behind a call for a slow code are good. Primarily, it is a way to spare parents the full, painful acknowledgement of the extent of their child’s deficits and the likelihood of his extremely poor quality of life or death. More importantly, it shields parents from having to make the painful decision to let their child die by choosing not to resuscitate or to stop treatment. It also protects the infant from the rigors of aggressive treatment that is likely to be unsuccessful.

It must be acknowledged that calling a slow code also spares the physician the helpless feeling of doing nothing, having to face parents with empty hands. He or she may also have a lurking fear of parents’ anger at the physician’s failure or malpractice charges if the infant dies with no medical interventions.

However good the intentions, though, calling a slow code raises significant ethical questions about deceit, paternalism, patient-doctor relationships, and teaching good communication skills.

What Is a Slow Code?

A full code or code blue involves calling a rapid response team and initiating appropriate treatment as quickly and effectively as possible with the goal of reversing an adverse event, returning patients to the status they had before the event that triggered the full code and restoring as high a level of functioning as possible. It is an emergency intervention with high priority, and speed is often critically important. A full code, properly executed, is often life-saving.

A slow code, by contrast, involves initiating some resuscitative measures but carrying them out slowly or omitting the most aggressive. Interventions in a slow code are limited in number, duration, intensity, or all three; for example, giving gentle chest compressions that do not crack the ribs. “Slow” also refers to the reduced alacrity with which staff responds to the call. The implicit hope is that the patient will die of his condition before they arrive.

In a recent article, John Lantos and William Meadow, both experienced and respected neonatologists who are well published in medical ethics, propose the use of a slow code as a legitimate response to situations like Dr. Holbert’s [1]. They define a slow code as a short-term trial of some intervention and emphasize that it is mainly a symbolic gesture, not expected to be effective but to give the appearance of doing something effective. Their article defends the use of slow codes.

Ethical Issues

Deceit. A slow code gives the appearance that something is being done that is expected to be effective, and the physician gives the appearance of believing that it will most likely be so. But the physician knows it is being done in a way that it is not expected to be effective. To put it another way, the physician has a hidden agenda; the goal is not the patient’s survival or improvement, but allowing the patient to die while somewhat protecting the family’s feelings. Thus, in action and in word, the physician is deceiving the parents.

Paternalism. By calling a slow code, the physician is making a decision for the parents according to his or her belief about the best interest of their child. The parents are thus denied their right, as decision makers for their child, to informed consent or refusal. One of the very basic tenets of medical ethics, in some places codified into legal regulations, is informed consent. Truly informed consent requires two things: that the decision makers be informed and that they give free, uncoerced consent. By leading the parents to believe that the physician expects the interventions to be effective, the physician withholds information that the parents would need to make an informed decision. Since they are making a decision based on incorrect information, it cannot be considered informed.

Patient-doctor relationship. Insofar as good relationships with patients and parents of patients are built on trust, the use of a slow code, by eroding trust, damages or destroys the relationship. True, parents may never come to realize that the “treatment” ordered was actually a slow code, but it is always possible they will figure it out. Even if they later recognize that ceasing treatment was a better choice for their baby, they are bound to resent that they were not told the truth.

Communication skills. What are physicians in training being taught when they are ordered to participate in a slow code? That doctors know best and parents are unqualified decision makers? That it is OK to deceive if your intentions are good? That the clever physician can find ways to avoid difficult conversations with parents, especially around life-and-death issues?

A Better Solution

It is our contention that physicians are led to use a slow code because parents are typically presented with a choice of two extremes: do everything or do nothing, i.e., do not resuscitate (DNR) [2] The problem is that for the physician, “do everything” means carrying out measures that are futile, interventions that are not expected to be of benefit and are likely to cause harm. Engaging in futile, possibly harmful measures is and should be morally unacceptable to them. For the parents, choosing DNR means giving up hope and choosing to let their baby die, which may be psychologically or morally unacceptable to them.

Viewing a slow code as a time-limited trial of some intervention or the use of some nonaggressive measures allows us to see it as a compromise or middle ground between the two extremes. It can be offered as a third option, one which provides an opportunity for the baby to respond if he or she can, but with the explicit understanding on everyone’s part that the intervention probably will not work.

The advantage of proceeding in this way is that there is transparency and no need for deceiving. The decision is made by the parents and fulfills both the ethical requirement for informed consent and the physician’s and parents’ need to do something rather than nothing. The physician must explain to the parents why doing everything is not a good option: it is painful for the baby, will not save him or her, and will leave everyone with regrets. Limited, less aggressive measures are appropriate if the physician thinks there is at least some chance of their working based on scientific evidence and a benefit-burden calculation, not just paternalistic judgment. If the physician explains his or her reasoning and actively recommends this third option and it is agreed to by the parents, then it constitutes a paradigm of shared decision-making. It also represents effective doctor-patient communication and preserves an honest, respectful, and rewarding doctor-patient relationship.

Terminology can be important, so we recommend calling the third option a limited trial run or limited resuscitation, dropping the term “slow code” with all the negative connotations we have described. When physicians hear parents say “do everything,” they should recognize it as a natural and understandable emotional response to hearing that their child is not likely to survive. But we argue that there is a better and more ethically responsible response than calling a traditional slow code.


  1. Lantos JD, Meadow WL. Should the “slow code” be resuscitated? Amer J Bioethics. 2011;11(11):8-12.

  2. Ladd RE, Forman EN. Why not a transparent slow code? Amer J Bioethics. 2011;11(11):29-30.


Virtual Mentor. 2012;14(10):759-762.



The people and events in this case are fictional. Resemblance to real events or to names of people, living or dead, is entirely coincidental. The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.