Balancing Prognosis and Excess Care

A normally reasonable, rational and responsible oncologist refused to see a patient with a diagnosis of cancer. She was available; in fact, the patient was just down the hall. She has a deep understanding of the particular disease. There was no problem with insurance coverage, precertification or co-pay. The doc is a provider on the patient’s plan. The reason was simple, as well as very complex. The patient is dead.

Do you know the old saw, “why do they put nails in coffins? “ It’s to keep out the oncologists. Nonetheless, in this case, the surgical service actually called the oncologist to do a consult in the morgue. Come on down, review the records, and, I suppose, examine the “patient.” I think we can assume that the verbal history would be brief. (?Patient not cooperative ?Patient does not respond to questions). Still, while this doctor is always willing to help, this request seemed to go beyond the pale.

The problem is this. The patient was admitted to the hospital with an acute surgical emergency. Labs were drawn, EKG performed, IV placed and he was stabilized. The patient was then whisked to the operating room, where, after an otherwise uncomplicated, routine and successful procedure, he died.

Now, the surgical service is asking a remarkable post-hoc question. Should they have taken the patient to the OR in this first place? Or, more specifically, what was the patient’s prognosis, re his advanced cancer? Did it justify the surgical risk? Instead of entering the OR, should the patient have entered hospice?

Any oncologist will tell you this is the sort of question we get all the time. Should a patient with Stage II lung cancer have bypass heart surgery? Is a diagnosis of CLL a contraindication to dialysis? Does a family history of cancer mean the patient should not have immunosuppressants?

However, there are several problems with the oncologist trying to predict the future and balance it against surgical risk. First, truthfully, we are not very good at it. Cancer doctors receive very little training in determining prognosis and data says we get it wrong much of the time.

Second, we are biased towards our patients and towards the potential success of cancer treatment. Not only do we sometimes believe too much in therapy, at times we are slow to tell our patients that the likelihood of the treatment curing or having a major benefit is small.

Finally, its not just cancer. There is the need to predict the future in patients with heart disease, kidney failure, advancing cirrhosis, debilitating AIDS and old age itself. For a challenge try to decide whether a patient newly diagnosed with Alzheimer’s disease should have a revision of an old pacemaker, a colonoscopy because of anemia, dental implants or hip replacement for moderate arthritis. Such decisions require an understanding of the course of the illness, which is often variable, the risk and benefit of the proposed surgery, which is not always evident, and, critically, an appreciation of the desires, goals and understanding of the patient.

It occurs to me, that in addition to doctors with primary responsibility for a particular illness, such as cardiologists, nephrologists, neurologists and geriatricians, there is a medical specialty that has a keen interest and specific training in prognosis. That is Palliative Medicine. Because of the need to decide whether patients are appropriate for end-of-life care, especially hospice, Palliative Medicine doctors have done extensive research on prediction, receive training in prognosis and a major part of the Palliative Medicine board certification exam is testing this skill.

Consulting not for care directions or quality of life assistance, but to give advice regarding the appropriateness of surgery, or other new medical intervention, would be an expansion of the traditional role of the Palliative Medicine service. It has the advantage of involving a clinician who has specialty expertise in prognostication, focuses on communication and has the perspective of a small degree of distance and objectivity around the particular case.

Palliative Medicine started, several decades ago, with a goal to improve pain control in the last few days of life. It has grown to offer a broad base of quality and support for patients and families who are at every point on the chronic disease path. Perhaps, prognostic consultation is a further way this critical specialty can help all of us, even we recalcitrant oncologists, to balance care and caring in patients with complex illness.


  • D Someya Reed
    This is a very odd post which may explain the lack of comments. I was inspired to do some research before offering my own. First, and to get it out of the way since you hit my "trigger word," hospice (as in "should the patient have entered hospice?" end of paragraph four). Choice of or entry into hospice is NEVER, EVER, EVER (do I need another "ever" for emphasis or is this enough?) the decision of a physician or anyone other than the competent patient (or the patient's family if patient is incompetent). Why? Because the patient MUST (with few exceptions) give up all curative measures. No medical professional, no administrator, no one has the right to make that decision for any patient! If any of them should decide for a patient then they are sentencing that patient to death; the ultimate and expected outcome of hospice. This is simple fact and is not intended to take anything away from those in hospices who truly and honestly attempt to improve the quality of remaining life for their patients. Back to this post. You set up a scenario of a refused prognosis which we find out is post mortem. You never explained why this oncologist refused? The whole scenario seemed to be suspicious, potentially litigious and, perhaps, seeking a CYA answer. Is this the reason for refusal? Next you stated that oncologists get asked this all the time? Did you mean requests for post mortem prognostication? Really? It doesn't appear so because then you discuss prognoses for obviously living patients probably undergoing some future treatment(s). This is where it's really odd because these two types of prognostication (one dead, others living) are not the same. I do want to note here how refreshing your honesty was starting with "First, truthfully, we are not very good at it" through "we are slow to tell our patients that the likelihood of the treatment curing or having a major benefit is small." This second part led me to some fascinating reading on "therapeutic privilege (or exception)" and the ethics of it or lack of ethics in the abuse of it. According to some medical ethicists, I read, abuse is rampant and some claim that therapeutic privilege should NEVER be used (and for the same logic I used regarding hospice above). So your answer to all of this was palliative medicine which you claim gets specific and extensive training in prognostication. Palliative medicine is not about cure but rather symptom relief. However, unlike hospice, it can be carried out concurrent with curative care but not likely by the same physician. Here's where I wanted to do some research because I couldn't see where palliative care practitioners would receive more prognostic training than solely curative care physicians. With as much as a pill prescribed by a curative physician there is a prognostication in answering the patient's question of when he/she might see results (from the med). Most prognostics are evidence-based upon collected data. It's still tough because each patient comes with too many variables to be truly comparable to any other. Palliative medicine, itself, only looks at the beneficial symptomatic effects in prognoses, not curative (that would be you guys). So essentially palliative care asks the prognostic questions, Will it give you better quality of life? (subjective) and Will it kill you? While oncologists, for instance, would answer both of these questions along with Is it likely to cure you? and/or Is it likely to extend your life span and by how much? I can't see then how oncologists should get the short end of the prognostic training stick. So, I checked most of the major medical schools and their course lists for palliative care training. There's not much. And, much of that is single day classes. Harvard Medical school had some palliative courses lasting a couple of weeks but didn't state how often classes met. Overall, prognostication was rarely, and in some cases never, mentioned as part of the course of palliative study. One exception was the ABIM (American Board of Internal Medicine). They offered exam blueprints and tutorials for their Hospice and Palliative Medicine exam. Their exam they stated was "developed jointly by the ABIM, the American Board of Anesthesiology, the American Board of Emergency Medicine, the American Board of Family Medicine, the American Board of Obstetrics and Gynecology, the American Board of Pediatrics, the American Board of Physical Medicine and Rehabilitation, the American Board of Psychiatry and Neurology, the American Board of Radiology, and the American Board of Surgery. I would hope that all of these groups would make a pretty substantial and thorough exam. They even have a table that shows the categories of study on the exam and a percentage for each category (totaling 100%, of course). The prognostication portion of the exam was listed as "Prognostication and Natural History of Serious Illness." TOGETHER these only accounted for 8% of the exam. Nothing on prognostication which I found was either extensive or a major portion of any exam. If this exemplifies the extensive prognostics training for palliative medicine students, as you stated and who don't have to concern themselves with curative or long-lasting effects, then non-palliative care medical students get scary little training at all. Something is horribly, horribly wrong and palliative medicine will not fix it. I still would like to know, however, why the oncologist refused to do the post-mortem.
  • D Someya Reed
    Just read this in (and quoted from) an Op-Ed piece in the LA Times. It's related in that palliative care is largely symptomatic care for the incurable (aka dying). "In four years of medical school, the average amount of instruction on death and dying is 17 hours." Consider this along with an article on doctors, PAs and nurses by Consumer Reports stating that "primary care doctors have about 23,000 hours of education and training." and again from the LA Times: "In 2013, only three of 49 accredited schools of public health offered a course on end-of-life care. Students do not learn more about dying, one report says, because death is a medical failure." So where and when are the palliative care (and hospice subset) physicians getting their training on caring for incurable/dying patients?

Leave a Reply