An article by Dr. Nancy Keating, from Harvard School of Public Health, in the journal Cancer discusses the issue of when physicians who treat patients with a terminal diagnosis discuss prognosis, preferences for extraordinary measures or a “do not resuscitate” (DNR) order, ideas about hospice, and where patients would prefer to have their death. Guidelines developed by the National Comprehensive Cancer Network recommend that physicians discuss these issues for patients who have a prognosis of less than a year, but what do physicians actually do? The relatively scant information on the subject suggests that physicians vary greatly in their styles.
To answer this question themselves, the investigators send out questionnaires that they received back from over 4,000 physicians, non-cancer physicians, medical oncologists, surgeons, and radiation oncologists. Specifically, they asked physicians when they would bring up the issues of prognosis, DNR preferences, hospice, and preferred location of death for a hypothetical patient with an estimated survival of 4-6 months but who felt well now. Options ranged from the time of the initial visit (”now”) to a point in the future when the patient was symptomatic, or there were no further feasible treatments, or when a patient was hospitalized, or not until the patient and/or family brought up these issues. The investigators also asked about physicians’ experience with managing terminally ill patients and how confident they were in their knowledge about managing terminal care issues for patients.
Overall, 65% of physicians reported that they would discuss prognosis from the beginning, with 15% on the other extreme of not discussing prognosis until or unless the patient brings it up. For discussion of DNR status, 44% favored discussing DNR preferences at the initial visit, with the remainder scattered across the milestones. Discussion of hospice was most commonly reported as something to defer until there were no further potentially effective palliative treatments available. Least likely to be discussed “now” was preferred location of death, which 21% of physicians saying they’d bring up from the initial visit, and 24% not discussing this until the patient brings up the topic.
There were also some differences among the physicians surveyed. Younger physicians were more likely to be proactive about initiating these discussions earlier. Medical oncologists and surgeons were significantly more likely to bring up prognosis than other physicians. Not surprisingly, the physicians who reported more experience and greater confidence about end-of-life care brought the issues up earlier than others.
The results are interesting and overall show a lot of personal differences, in keeping with the evidence as well as what I think most of us see in the real world. Some physicians feel that this is part of appropriate care for terminally ill patients, since studies show that physicians are not especially good at assuming correctly the preferences of a patient for issues like DNR status. Other physicians are concerned that discussing end-of-life care dispels hope and sets a negative tone and evade these issues as long as possible.
I personally feel that it’s important for patients to have some real understanding of their prognosis if they are really going to be equipped to make an informed decision about treatment. Someone who is agreeing to start a potentially challenging regimen of two or three agents for advanced lung cancer should know the limitations of what treatment can offer, and specifically that we aren’t giving it with an intent or realistic expectation that it will be curative treatment. On the other hand, I’ve also become more nuanced and try to read my patients in terms of what they’re receptive to discussing, compared with being more universally proactive in my earlier years in practice. And even though I’m definitely among the oncologists at my institution least likely to avoid these discussions by the time they become pressing issues, I’m certainly not someone who routinely discusses hospice and a patient’s preference for where they would prefer to have their death occur on an initial visit.
But is that wrong? My hospital now requires that we specify a preference for “Full code” or DNR for every patient admission, but I know that we physicians don’t always have these discussions with patients in whom they don’t envision this to be an urgent issue (and they often still don’t when it really will be an immediate issue). The guidelines suggest that doctors initiate these discussions from the very beginning if a patient has a prognosis of less than a year, even if they feel well. I personally don’t feel that I should, based merely on guidelines from an expert panel that are based on no particularly strong evidence that this is more helpful than waiting until you’ve established a relationship with a patient and end-of-life issues may be relevant in the near future. Some of the physicians who responded that they don’t pursue these discussions in an early visit probably feel the same way, though I’m sure other physicians just don’t want to have a discussion that can be awkward and, in some patients, may alienate or offend them.
Over time, I’m coming to feel that our approach to these issues should be individualized, just like we individualize our medical approach based on patient health, performance status, molecular markers, preferences about aggressiveness, and many other factors. Some patients may take longer than others to reach a point when they can discuss these issues, and I think that’s OK and even quite appropriate. To me, using guidelines to direct the timeline for having these discussions is too “color-by-numbers” in a time when people need individualized management.
I think it’s very important to hear from people on the other side about what patients and caregivers would be receptive to, or whether they’d prefer to bring these issues up and not have these topics initiated by their physicians.
Posted on January 17, 2010 at 11:44 am
As a care giver, I don’t mind doctors bring up this issue at the very beginning, even the odds are pretty bad and brutally against my loved one. I remember the first time our doctor telling me that my Mom will likely not survive longer than a year, I can still maintain my upbeats and feel full of hope. Now looking back, I think my confidence comes from my newly learned knowledge of the disease, and the way the doctor deliver the information.
I’d say that timing of this kind of discussion is not that critical for a care giver. It is an ongoing issue. With ups and downs down the road of treatments, there are many times when we have to face it. What matters most is the way we approach this issue. I would appreciate more articles on the arts of discussing prognosis and death with a cancer patient.