Monday, January 11, 2010
Journal of Clinical Oncology has a paper on cancer patients' attitudes about discussing advance directives. The study is posited as a follow up to one published 10 years ago which suggested that cancer patients did not want to discuss ADs with their oncologists. The data come from a prospective study involving all patients consecutively admitted to an academic hospital's inpatient oncology service over several months; patients (N=75, mean age 51 years, 23% African American; the majority had hematologic malignancies) were interviewed about AD completion and preferences.
41% of patients reported having completed an AD (all were given a description of what was meant by an AD); in multivariate analysis older age was the only predictor of AD completion. 75% of the patients had discussed an AD with someone - mostly family (23 patients reported a discussion with a physician). Only 5 patients reported a discussion about ADs with their oncologist. When they asked patients if they would like to discuss ADs with their oncologist, only 23% said they would (this is similar to the finding in the study 10 years ago). When, however, they asked patients which, of all their doctors, they would prefer to discuss ADs, a plurality said their oncologist (48%). That is, if they have to do it, they'd prefer their oncologist. Notably the vast majority of patients (87%) thought that physicians admitting a patient to the hospital should ask about ADs (they indicated this was not only ok but an important thing to do). Thus, the title of their article, and this post - 'Paradoxes in ACP....'
(They also asked patients about knowledge of hospice care and palliative care. 21% of patients reported knowledge of 'palliative care' vs. 81% for 'hospice care,' and hardly anyone said they knew anyone who had received palliative care.)
Basically these patients, most of whom did not have an AD, weren't particularly interested in talking about them with doctors (oncologist included), while recognizing that when you are admitted/acutely ill it's good to bring it up with a stranger...but that if they had to discuss them with a doctor most preferred their oncologist. This is summarized by the authors as: "while most patients would not like to discuss ADs with their oncologist, they would prefer to discuss them with their oncologist."
I've been mulling over these findings for a couple days now, trying to make sense. I wonder whether the fact most of the patients had hematologic malignancies was a factor; while many of the patients had a low likelihood of long-term survival, it is still likely that most of the patients at the time of the interview were expecting long-term survival/cure and whose treatment intent was essentially curative (as opposed to patients hospitalized with advanced solid tumors who, one hopes, understand their cancers are incurable). I think having more solid tumor cancer patients would have attenuated the findings, but not necessarily changed the paradoxes.
In some ways it's a testament to the importance of how questions are worded in research (as in real life): Would you like to talk with your oncologist about ADs? is a different question than Which of your doctors would prefer to talk with about ADs? (This is the Journal Club teaching point of the paper.) Granting that difference, however, doesn't mean that these patients resoundingly endorsed routine AD discussions in general.
We've expressed some AD skepticism on the blog before (e.g. where's the data they do anything), and I sometimes wonder when reading research like this that patients view ADs as some sort of abstraction and don't really understand why us clinicans care about them so much. Which is not to say that most physicians do care about them much. Part of what I'm saying is that for us as doctors to tell patients they should consider an AD we also need to be talking with them about their expected future, and that the AD is less important to patient care than that actual discussion about what's going to happen, or what will happen if consolidation chemotherapy for your acute leukemia fails. It's those discussions, and not the recommendation to do the paperwork, that I *think* improves care (and indeed the authors cite some of the Coping With Cancer Study outcomes - that study showed improvements in care associated with EOL discussions, not per se with AD completion). I wonder if patients kinda get that too, and if they've had some 'AD discussions' which were along the lines of 'you really oughta fill out this paperwork' without the context of the broader talk or at least broaching of the topic of prognosis/expected course, they don't really report the need to engage with their oncologists about the paperwork...?