The situation
In medicine generally, but critical care especially, we often care for patients at the end of their lives. Sometimes, we succeed at prolonging their life, and that pleases us, because we like to fancy ourselves healers. At other times, we are unable to significantly bend the arc of mortality, such as when a chronic terminal illness approaches its final stages, or when catastrophic acute disease overwhelms our ability to manage it.
At such times, after recognizing the poor odds, we may believe it’s best to stop fighting a losing battle. Although many patients and their decision-makers will assume, by default, that we will (and should) keep attacking the disease in an effort to cure it, as clinicians we often suspect this will be futile. We cluck our tongues at each other and say things like, “This is not going to end well,” “Why are we still doing this?” and “Is this really what they want?”
Eventually, we take the family aside for a “family meeting” to discuss “goals of care.” This is a euphemism for a quiet, earnest conversation in which we will try to convince them to set limits on our curative efforts, and transition towards more comfort-focused care—a new approach which will be unlikely to prolong life, and may even shorten it.
We frame this as a decision to be made solely by the patient or their decision-makers, and it generally is, at least in the United States. (In other settings and other countries, however, the decision may be made for them, with the rationale that ICU beds are not infinite and cannot always be exhausted on patients with a poor prognosis.) We try to avoid paternalism by dictating the “right” answer; instead, we ask what the patient would want, based on their personal values, irrespective of our own.
However, we do have our own beliefs, and as a rule, we believe that in such desperate situations, aggressive curative care is “bad,” and palliative care is good. This universal truism is accepted by most clinicians, woven into the culture of our teams, and transmitted to trainees as part of their invisible curriculum. Polls of physicians, for instance, describe an overwhelming majority who would personally wish to be made “do not resuscitate,” in contrast to the bulk of their patients who are generally being resuscitated.
There is a perception that as medical providers mature, particularly in high-acuity resuscitative fields like critical care, they transition from a love of “saving lives” toward a more sagacious attitude that accepts the things they cannot change. For many, successfully making a moribund patient “comfort measures only” is nearly as satisfying as the rare “save” who (with our help) manages to cheat death and return to their former productive lives. Indeed, an excessive focus on life-saving is considered somewhat childish if not actually harmful, since it creates cognitive dissonance and “moral injury” when contrasted against the large number of patients whose lives cannot be saved. This has been true since before the burned-out interns in The House of God struggled to find meaning in their wheel-spinning care of moribund gomers.
Do patients define their own goals of care?
In short: yes, but we wish they didn’t.
We ask what they want done, but it isn’t a true open-ended inquiry. It is a test, with a right answer and a wrong one, like the attending who asks an intern to describe the Krebs cycle, or an interrogator who demands to know if a suspect is guilty. Often we won’t even ask the question if we assume we should pursue aggressive care; when admitting the healthy 30-year-old with appendicitis, how many of us confirm whether they would want to receive CPR? Like a poorly-applied diagnostic tool, that question seems to fall below a certain test threshold: we are so convinced we already know the truth that an answer would either be redundant or erroneous. Only rarely will a patient or family member favor less aggressive care than we deem appropriate; when it does happen we are thrown for a loop, and may to change their minds. (“But there’s still hope!” we cry.)
Thus, full-bore disease-limiting care often occurs because of patient wishes but despite our own. We pay grudging lip service to patient autonomy, but maintain our own beliefs about what’s right—who wouldn’t?—and behind the closed doors of our team rooms or our own minds, we suspect such choices are simply wrong. People might reasonably disagree in preferring chocolate over vanilla, or their favorite genre of music, but—we think—could only want “everything done” in such dire circumstances due to ignorance (not understanding the prognosis) or denial (understanding it, but not accepting it).
This perspective would make sense if prognosis were a sure thing in every case. If it were 100% clear that the patient will not survive their acute illness with an acceptable quality of life, then there would be no rational reason to continue treatment. But even the most arrogant among us would never predict an outcome with 100% certainty. We merely play odds. And from the wrong side of those numbers, a 10% chance, or a 1% chance, or a .1% chance might sound pretty good when the alternative is 0%.
So why do we tend to feel that rolling those dice is the wrong decision? Perhaps because the harms of aggressive curative care often seem to outweigh the potential benefits. We’ll discuss this more in Part II.