Friday, May 17, 2013
A better POLST (Physician's Orders for Life Sustaining Treatment) and informed consent for resuscitation: can we do this better without "playing God" ?
The United States has come a long way in the last 2 decades since 1991 when the Patient Self-Determination Act was instituted. In most hospitals patients are asked what their wishes are regarding resuscitation and many states have instituted POLST (Physician's Orders for Life Sustaining Treatment) forms which spell out which interventions are acceptable to individual patients when they are very sick, things like blood transfusions, antibiotics, feeding by nasogastric or gastric tubes, intubation and ventilation and chest compressions and electrical cardioversion in the event of cardiac arrest. Theoretically we discuss these things with every patient when they come into the hospital, but we don't really do it very well which results in misunderstandings and needless misery.
When the patient is competent to discuss life sustaining treatments the conversation often goes like this:
Doctor: What would you want done if your heart were to stop? Would you want us to do chest compressions and put a tube down your throat so we can support your breathing?
Patient: (thinking, "Why are they asking me this? Am I gonna die? Of course I want them to save me. My heart could stop right now! What about my grandkids?") Sure I'd like that. Just give it a try. But I don't want to be kept alive if I'm a vegetable.
Doctor: (thinking, "I sure don't want to break all of this nice guy's ribs and traumatize his upper airway and, if we did get his heart started again, have him helpless in the ICU while we try to decide whether to put a feeding tube down and send him to a nursing home. With any luck he'll die in his sleep when he's not in the hospital.") OK then. We'll put you down as full code. Just sign here.
I went online to see how other countries do this, and found that there is a great deal of variation. In many middle eastern countries resuscitation is performed on all hospitalized patients in the event of cardiac arrest, regardless of patient or family preference. In Australia there are many different policies which are different in different hospitals, and may or may not involve patient and family participation in decision making. In England the patient is involved in these discussions if he or she is competent, but if not, the doctor makes the decision, and if the family has an opinion this is taken into consideration but the doctor really has the final say. I read an online discussion about an 86 year old nursing home resident in the UK with heart failure who was admitted to a hospital and suffered a cardiac arrest. He did not have CPR due to a nurse thinking he had a DNR order, which he did not. After the event it was agreed that CPR would not likely have saved him, but it was the discussion after the article that was particularly interesting. About half of the people said that it was terrible that mistakes like this ended in a patient not getting life-saving CPR and that in many cases doctors made these decisions and "played God." The other half were people who said that they had seen CPR on old people and it was brutal and hardly ever worked.
Clearly doctors should not get to or have to "play God." Clearly, also, people who we ask to make these decisions, patients or families, usually don't know what we are really talking about when we ask them to make a decision about resuscitation ("code status.") Physicians are much less likely than our patients to request resuscitation in the event of cardiac arrest. I think our values are pretty much the same as those of our patients, we just know more about what cardiopulmonary resuscitation and its aftermath really look like.
So here is a good solution that preserves the autonomy of the patient while allowing the physician to do his or her job, which is taking care of the patient while following our Hippocratic Oath to "do no harm".
Cardiopulmonary resuscitation (CPR), intubation and ventilation are medical procedures, just like a tonsillectomy or a hip replacement. There are expected risks and benefits. The risks are not insignificant and the magnitude of the expected benefits vary with each patient. If a patient really wants CPR, he or she or a surrogate decision maker should sign a consent form (well in advance of experiencing a cardiac arrest) which includes risks and benefits. The benefits, of course, would be to live and recover enough to be able to leave the hospital. In the least complex patient, 1 in 5 of those who have a heart event which requires CPR can be expected to survive to hospital discharge. In the most complex, the expectation is 1 in 20 or less. The risks include, but are not limited to: pain, broken ribs, punctured lungs, trauma to mouth, teeth and upper airway, aspiration pneumonia, loss of brain function, multi-organ failure, prolonged dependence on caregivers including being confined to nursing home, monetary expense, and all the usual complications of prolonged intensive care unit and hospital stays. Also, should we not be successful, loss of the opportunity to die in peace.
Presently our POLST forms say none of that. Different states have different forms, but I have been involved in completing ones in California, Oregon, Idaho and Washington and they share similarities. They start with a check box for CPR (cardiopulmonary resuscitation) or DNR (do not resuscitate, allow natural death.) Then there are other boxes specifying the general level of interventions, including comfort care (giving only treatments that provide comfort without attempt to prolong life), limited interventions (lengthen life, but generally not in an intensive care setting and not using advanced life support techniques) and full treatment (everything, dialysis, ventilator, electrical cardioversion, the works.) Then there may be questions about whether artificial nutrition by IV or tube is acceptable, whether blood products are OK, whether antibiotics should be used. It is really hard for a patient or family to get through all of these questions, and many of them are too complex for a person without healthcare experience to understand. Still, they are a good starting point for discussion.
In general the patients who I talk to just want me, as the doctor, to make the right decisions at the right time that are most likely to get them what they want. Most of them want as much good life as possible, the relief of whatever discomfort or disease brought them in, and to have this done as quickly and economically as possible so they can go home. We have no box for this on the POLST form.
I would propose that we should add this box, up at the top, as an alternative to "CPR" vs "DNR." I would propose that the choice be called "Resuscitation at the discretion of physician, guided by my goals of care." This would allow a nuanced decision about resuscitation--from none at all if the patient had been declining and not responding to treatment and was found unresponsive and pulseless (a situation which almost never results in success) to full resuscitation for a witnessed collapse with ventricular fibrillation in a patient who had a pretty good level of function.
This does, however, require another piece of paper, which I think should be part of any patient's admission to the hospital, and probably part of the chart at the primary care doctor's office. This would be "goals of care." There was an article in the New England Journal of Medicine last year that talked about altering our ideas of success in medical care to reflect how well we helped a patient achieve their goals rather than focusing on specific markers of disease control. Some patients value not being dizzy and not taking a bunch of expensive pills more than they value good blood pressure control, for instance. I think it is important to know, at the time of hospitalization, what a patient really wants. Do they really need to be discharged by a certain date or time? Do they have lousy insurance and need their care to be as thrifty as possible? Do they really value pain control, or sleep at night, or making sure to have a certain meal on time? Do they need quiet, or visitors? Do they want to avoid antibiotics or medications that can cloud their thinking? Do they think that spending time in a nursing home would be OK? Are they hoping to die in the hospital because their burden of disease is becoming intolerable? Do they want to make it home for Christmas or live until a grandbaby is born? We don't ask these questions and we should, or at least somebody should.
If a person chooses "CPR" or "RDP" (resuscitation at the discretion of physician) they, or their surrogate, really need to read and sign the informed consent form for resuscitation, because they do need to know what this means.
I hate filling out forms, but if they help focus treatment so it is more appropriate and if they help me understand my patients and communicate with them more effectively, I'm OK with that.
So... bottom line: In order to have patients make better decisions about resuscitation, we need to share more information with them and allow them to depend on our clinical judgement to help them have the outcomes they really want. To do this we should: 1. Add a box to the POLST form that specifies "resuscitation at the discretion of physician" as an alternative to "CPR" or "DNR". 2. Create an informed consent for resuscitation which makes clear expected risks and benefits and have everyone who wants resuscitation sign it. 3. Make sure that patients are asked about their goals of care at the time of admission to the hospital and that physicians read them and honor them as much as is practical. 4. For those unable to complete a POLST form or give informed consent for resuscitation, physicians should make the decision about whether to resuscitate based on most patients' goals of care and good clinical judgement. This is, after all, what we would do for such a patient with any other procedure that we perform in the hospital.