skip to Main Content

I think we all probably agree that health care providers should give patients with a life limiting illness the option to discuss their prognosis. But what about discussing prognosis with those patients who may not have any particular life limiting illness, but have just lived a long life? Are we as forthcoming about prognosis in this population as we are with a population of patients with advanced cancer? Should we be?

In this week’s New England Journal of Medicine (NEJM), Alex Smith ventures outside of the blogosphere to give his perspectives on discussing prognosis with very elderly patients who may not have a dominant terminal condition.  Along with Brie Williams and Bernie Lo, Alex lays out in this perspectives piece the importance of and the barriers to having these types of discussions.

I won’t recap the article here, except to restate the overarching recommendation that clinicians should “routinely offer to discuss the overall prognosis for elderly patients with a life expectancy of less than 10 years, or at least by the time a patient reaches 85 years of age.”

The field of palliative care has come a long way in getting physicians to discuss prognosis with critically or terminally ill patients. Kudos to Alex for highlighting the importance of doing the same for the vast majority of elderly patients who fall outside of these nicely defined groups.

Questions though remain and the expertise of both the geriatrics and palliative care community are needed to sort through them.  Some that come to my mind include:

  • How should we determine prognosis in this group as just age is too blunt of a tool.  Should we use prognostic indices, clinical judgment, or a combination of both?
  • How much confidence do you need in a either a prognostic index or clinical judgement for it to change management on clinical issues like cancer screening?  If an 85 year old has only a 25% chance of living 10 or more years, is that too low to recommend colon cancer screening?  
  • What should we be doing differently in those patients that are identified as being at high risk of dying in the next year versus the next 5 years (a common scenario in the oldest old)?
  • How do we make these discussions the norm and not the exception?

I’d love to hear your perspective on Alex’s piece.  Please comment below.

by: Eric Widera

This Post Has 12 Comments

  1. Is this aimed at healthy old-old people? Are doctors making these kinds of decisions now without discussing them with their patients? Is the point to bring this out in the open? Or, is the goal to limit preventative medical care because, well, they're just too old?

  2. Hi S, this is about empowering very elderly individuals to make informed medical decisions and life choices. In the paper, we cite research that suggests a majority of very elderly individuals want their doctor to discuss prognosis with them. People want to know for a variety of reasons. Most very elderly people are aware that they're in life's final chapter, so this discussion would be "bringing it out in the open," as you suggest. Doctors and patients can then work together to set priorities in the context of the life expectancy.

    Thanks and interested in other people's comments and questions. If you'd like a full text version of the article, you can email me at aksmith at ucsf dot edu. NEJM sent me an electronic copy for distribution for noncommercial use (GeriPal isn't commercial).

  3. I'm not quite sure that I agree with what he writes here. In elderly patients that have no specific life-threatening illness, it is quite hard to give any reasonable estimate of life expectancy (I'd be glad to see a study that shows otherwise). I presume that elderly patients realize that they will eventually die – they see their friends dying and also know that they are slowly declining. What purpose does giving a very non-scientific prognosis have?

  4. Thanks Moshe, this is an excellent point. Prognostic tools exist to help clinicians estimate long term prognosis. See GeriPal contributor Sei Lee's JAMA article from 2006, a 4 year prognostic index for community dwelling elders. 82% of the time, the index sorted patients who died within 4 years from those who did not.

  5. Some older folks are quite hardy which is the case in my mother's family. They all live independently well into their 90's and usually die suddenly of a massive heart attack or stroke. In fact my great grandfather plowed a farm field the day he died– you might say he died with his boots on. It doesn't seem right to me that people like this should be advised, encouraged to have less medical treatment than say a 60 year old with COPD and congestive heart failure.

  6. Great article Alex!

    Sara B, I think you highlight precisely why screening may not be useful. If the people in your mother's family die suddenly of a heart attack or stroke, than what good is a mammogram or a colonoscopy? The concept that older adults have competing morbidities (such as massive heart attacks and strokes) are precisely why many of these tests become less beneficial as life expectancy shortens.

    Would you do a colonoscopy on a hardy 101 year old?

    I just love the nuanced work on cancer screening that has come from UCSF over the past several years.This is not ageism, it is just good, nuanced thinking from people who understand the changing risk/benefit ratios as people age.

    The challenge, as evidenced by some of the posts, are that these nuances are counterintuitive and very difficult to articulate.

  7. The question is not should we limit care, it is whether or not we give someone the choice to do so. When I discuss adavance care plans with patients, they are often not aware of the evidence and their "ability" to decide on scientific grounds to forgo screenings. Most of them are well aware that they have few years left, but are seldom given the opportunity to discuss this reality.

  8. As a geriatrics fellow, this article helped me think about how I am providing and framing the care I offer many of my patients. For me, the important piece is how we communicate around a limited prognosis due to multiple comorbidities when there is not one thing to hinge the conversation on.

    A few days ago, I talked to a patient who is 84 with CKD, diabetes, CHF, worsening functional status (and more) after a CT scan. As had already happened twice before in the last month, it showed nothing menacing, but found another possible problem in a different organ. Of course, the radiologist suggested another CT scan that would require a large contrast bolus. The question we confronted was how far to go with serial incidental findings. He wanted to proceed and didn't see why we would stop. As I sat with him, I tripped over how to properly frame the risks that keep mounting of all our workups and the "big picture" of his overall prognosis.

    I really like the suggestion that Dr. Widera and the authors make of normalizing these discussions and providing some examples of how to steer the conversation. Very helpful. I would be glad to see follow-up articles with even more specific communication tools/strategies.

  9. At age 86, my Mom is very interested in medical interventions that can provide short-term improvements in quality of life, and rejects strongly any screening that might lead to highly invasive interventions like surgery or chemotherapy. It is clear that her doctors are very uncomfortable with this. I hope that, as a society, we get much better at clarifying, and then accepting, individuals' goals.

  10. I could not agree more than 80% of patients want to discuss this with their doctors they are just waiting for them to initiate the conversation

    I work with a very diverse population in San Mateo County mostly Phillipino, Mexican ( and other latino population), Caucasian, Russian and other ethnicities. As a physician with a focus on geriatrics – many patients will tell me that they are happy that they can have open honest communication with their physician. They want to discuss the potential benefits of a treatment or evaluation such as a colonoscopy, CT scan with contrast dye, utility of certain treatments such as ICU levels care with different invasive treatments.
    The key is not to place one's own value judgments, but rather explore the patient's own value judgments and thought processes. If asked for recommendations, it is ok to make recommendations while checking with the patient that your thought process for the recommendations is in agreement with their own values.
    It is also is important to provide correct information so they are not making assumptions based on the ideals of watching Drama tv shows where every very sick patient survives CPR, and every patient who was sick in the hospital from a stroke wakes up and returns miraculously right away to their previous levels of function without prolonged rehab, or complication along the way. Not to mention that hardly ever will a frail patient be able to return to 100% of their previous mental or physical function.

  11. Unless overtaken by dementia, most older-older people have some idea of their likely end date, based among other things, of what they know of their own parents,etc, and their own feelings about their own growing fragility, or lack thereof if fortunate. With a mom who's 89, she is aware that she is likely to be gone within about 3 years. However, doctors really should be clear with patients about the dangers of CPR, for instance; and the lack of usefulness of many other tests and procedures. Minimally, all patients should be made aware of in what way any procedure will inform further treatment; if a test is done to determine the need for treatment that need not, or shouldn't, be given, what is the point? As in all doctor patient communication, the issue is about clarity and honesty.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top