Tip:
Highlight text to annotate it
X
So I want to share one more example.
I want to thank the four speakers and Steven for really
thought-provoking thoughts, a really nice summary and meta
thoughts about the talks. I don't really have
anything to add. I mean I think most of it's been talked
about and we've taken a look from the high view.
What I'd like to do with this example is I'd like to pull us
down to the ground level view.
And rather than me pushing information out at you,
I'd like to put the example out in the room and have us just talk
about what would you approach be in this case.
So I'd like to tell you one story from two perspectives,
and then have us enter into a discussion about how we
would approach this guy. I'll ask Bob and Rick to be
ready with the mikes as soon as I'm done.
So this is a patient case.
I'm going to tell it first from the perspective of the doctor and
then I'll tell it from the perspective of the patient.
This is a real case, by the way. Mr. Renny is a 49-year-old
veteran with a history of chronic pain, schizophrenia, substance
abuse, prostate cancer and Type 2 diabetes.
He has been receiving care from me for the past two years.
During that time, his diabetes has remained under poor control.
When he originally came to see me, he was on no medications for it.
Over the past two year he's been on increasing doses of oral
hypoglycemic medications. He's now on maximum doses
of metformin, glyburide and rosiglitazone and still has
a glycohemoglobin of 11.2. That's really bad.
He states compliance with all of these medications.
Current clinical guidelines suggests starting insulin
in such cases. These guidelines are based on strong evidence
that reduction of long-term complications of diabetes
can be achieved by optimal blood sugar control.
Given Mr. Renny's poor control on maximum doses of oral medication,
I think he should begin subcutaneous insulin injections
twice a day to be titrated to optimal glycemic control.
He says, "I'm 49. It's been a hard life.
The voices first started talking to me in 'Nam back in '74.
At that time, my friends were all using heroine.
I started too. I got 15 years of addiction behind me.
I've been clean now for 12 years.
Things got really bad when I got that prostate cancer.
They did that surgery that left me with pain in my leg that
never goes away.
The voices are angry about that, but I try to ignore them.
Mostly, I try to live a straight life and get by.
The doc says my diabetes is getting bad.
I try to take the medications when I can get a ride to the
VA to get them. They don't bother me too much.
I'm good at remembering to take them.
The doc is happy about that.
I'm worried about this next visit, though.
He said last visit that I might have to start shooting myself
with insulin. I don't think that's a good idea.
The voices say that if I start shooting insulin,
I'll start using again. I don't want to go back to that life."
So let me open up. How are you going to approach
this guy? What are you going to do? What are you going to say?
What are you going to think?
Dominick Frosch: I guess I would start more with a question about
the doctor's exploration so far which is one is he aware of
the substance -- is the physician aware of the substance
use history, the gravity of it?
And has he had a conversation with the patient about his linkage of
shooting something is like going back to IV drug use?
Paul Haidet: So for the purpose of this discussion,
I presented it as if the physician wasn't.
But let's for the purpose of this discussion,
let's assume that we have elicited the patient's preferences.
Let's assume that the patient has come in and and the doc says.
"I think we're at the point where we need to start insulin."
And the patient says, "Doc, I don't think that's a good idea."
And the doc says, "Well, why not?" And the patient tells
this story about shooting insulin. So let's assume that all of this
baggage is out on the table for both people to see.
Liana Fraenkel: So I think I would think about this for
a while. It's a good thing I'm not a primary care doctor.
But we see this frequently at the VA.
So injection issues, substance abuse issues are a big problem.
Some patients tend to have particular well-formed ideas
about the voices they hear.
So in some cases, they can't shoot themselves,
but a nurse could shoot them or somebody else could shoot them.
So in this case, again, when there is clearly major benefit to be
derived and major harm by sticking with the current situation I'd
explore all potential avenues that might not be-- have the same
response to those voices even one outside of the box to see.
Clarence Braddock: I was struck by a couple of things.
I was really struck, Paul, by the comment about the patient that my
doc's really happy about that.
And there was something about that that would make me, again,
what would I actually do?
I think I would want to explore with the patient what his view
about the diabetes, under the broader heading of understanding
the illness experience and what the diabetes is to him on the
priority of life concerns, how he conceptualizes that.
Because that comment made me think there may be at some level in
which the patient is constructing the diabetes as something that the
doctor's worried about and he's doing it for the doctor.
And the issue of adherence, viewed through that lens, it's like,
well, I'm doing a pretty good job of making the doctor happy,
but I'm not necessarily constructing it as how am I
helping myself to be healthy.
So I guess that would be the first thing is I didn't hear in the
story or the premise about the interchange about how much we've
explored what his construction of diabetes is.
And how maybe working with that together might yield some fruit.
I mean because I'm still not clear if the patient is maximally or
optimally adherent.
I guess the second comment I'll make is around goals.
And, I think, the 11.2 that was a good device by you because I think
all of the clinicians an AIC 11.2, you didn't make it 8.2 because if
it was 8.2, again, if I had that conversation with the patient,
there are situations at which the goal of tight control on older
patient or co-morbidities, but 8.2 actually might be too tight.
But even at 11.2, I think, we'd want to reach some shared notion
of what is the goal. What do we want to get to?
It might not be eight. It might be ten.
And, again, I think, that's where the notion of practical wisdom.
I forget who put that out there.
But, again, the notion that the guidelines have to be adjusted in
the sense of goals that connect to reality.
Laurence McCullough: Just some historical context.
My understanding is that from talking to people that serve there
is that drug use, including injectable drug use by soldiers in
the combat areas of Vietnam was widely tolerated by the
officer corps.
And so might he associate that ironically with a kind of
abandonment by his own doctor.
And the other is whether he's flashing back and there's some
undiagnosed PTSD going on here as well because that he used the
phrase shooting himself carries a metaphor of weaponry.
And so I was struck by that.
So I just wonder what you, as clinician would see as trying to
help this guy, he would actually see as yet another superior
person, authority figure is going to abandon me to my own devices.
Tony: So my approach would be to build on the things that have
said. I think my sense would be it's only in the context of my
proving to him that I understand his situation and his fear about
this, that I could offer some other kind of way forward.
So actually the point that Leana brought up earlier about the
construction of the preference, actually,
would have to do with the-- I would build on the construction of
this story and this anecdote and his willingness to tell me.
And then to add a piece to what Clarence was talking about,
after the exploration of what it means,
I think the question I'd be asking myself as a clinician is can I get
to a point where I could offer some kind of reframing of this
issue in a way that would take it out of the realm of his old
injection stuff.
You know, could I get to the point in the conversation where I'd say
I noticed you used the word shooting and for this I would
consider it like a new kind of medicine and this is like an
injection of insulin.
And so I would see if I could reframe it in a way that would
offer as a way forward.
And to build on ways that he has been successful in caring for
himself because clearly this is a guy who has overcome a great deal
of adversity to be functional.
And so could I link his new behavior that I'm trying to get
him to towards all of the stuff he's doing to get through
schizophrenia, recover from his experience in Vietnam, et cetera.
And in that sense, there is a decision that comes up,
but it's more that I have to build the right context for him.
And in a way, it is kind of a serious-- maybe it is a kind of a
series of nudges around that. Paul Haidet: How strong is the
pull to get this guy on injectable insulin? How strongly are
you feeling that pull? Ronald Epstein: I guess I
would be wondering if this is the right question to be asking.
Because if you're looking at his overall survival and
morbidity things like exercise and smoking and blood
pressure control and lipid control may actually have more of an
influence than a few points. The second question I'd be
asking myself is even if he agreed to use insulin,
what's the best possible outcome of that use? I mean this is not
the kind of guy who is going to have an A1C of six on insulin.
And will reduction by a point or two really make-- so it's very
easy to hang on to a number.
It's kind of very easy for us to choose that because it's a linear
scale that's more easily measurable.
But it may actually not have as much impact on his life as a set
of smaller decisions about less controversial issues.
So I guess I would probably try to explore all of those to the degree
that I could before going with what seems to be the
most difficult one. It may sound like avoidance, but I just want
to make sure that every stone wasn't left unturned.
Paul Haidet: Well, I guess, what I'm hearing you say which I think
this case also brings up is what is the outcome or outcomes we're
interested in here? Is the outcome a certain number
of risks for cardiovascular events in 10 years?
Or is the outcome the chaos of a life back on heroine?
What outcomes are we dealing with and how do those outcomes play
against one another?
Mary Politi: I was also struck the absence of the lifestyle--
conversation about lifestyle recommendations.
And the physician described the case in a very medicalized way,
totally differently from how the patient described the condition.
And there was nothing about the patient's emotions about this.
You know, mentioned schizophrenia and chronic pain as if they were
just checkboxes of his medical history and didn't mention the
heroine use if he knew about it.
And it was very focused on this one concept and I think that
probably could get glossed over in the patient interaction when
they're discussing the actual diabetes.
Whereas in a lot of Victor Montori's work with the diabetes
medication cards, all of a sudden when you just start asking people
about what it is that they prefer and why this patient may be
resistant to insulin all of those other issues may come up and they
may have come up with some other idea about how best to manage
diabetes like diet and exercise.
Some things that may help the chronic pain as well or may help
some of the other conditions as well,
or the prostate cancer recurrence and things like that,
that perhaps the patient would become more invested and on board
with that if he felt that he was being understood by the clinician
in more than just a medicalized way.
Paul Haidet: Kind of like a collision of worlds,
which is what happens in that examination room.
Tony: I think the big thing is to...
Paul Haidet: ...keep that relationship going.
Tony: Yeah.
And that he stays at some level of care.
If he stayed at some level of care,
a guy with a million problems like that,
I mean that would be more important overall to him than
improving one of those numbers. Paul Haidet: A friend of mine,
David Buck, is the medical director for Houston's
Healthcare for the Homeless.
They did a project where they were interested in goal setting.
They were going out to the underpasses around Houston.
They were asking patients what their their healthcare goals were.
Well, in Houston if you don't have insurance you can go to the county
hospital if you have a "gold card."
But they change the rules often and it's hard to get one of these
cards. So the goal wasn't, "I want to get my blood pressure
under control." The surprise to them was, "I'd just like to get
help getting a gold card." That's when you step back and say,
okay, that changes everything. Now I'm seeing a completely
different world than the one I am living in.
Dominick Frosch: Just briefly, all of the points made really
resonate. In particular, I think the point that Ron raised
regarding what we are going to achieve here if we lower the
A1C by one point --increase the risk of hypoglycemia,
increase the risk of relapse, and so on.
I also want to just briefly come back to the point that Clarence
made about ownership of the disease because, I think,
it says something about how physicians approach this
with their patients.
Last year I went to my 23rd annual eye exam.
Now, I've managed my diabetes very tightly for many years.
And so after the exam the ophthalmologist who'd I only seen
for the first time that day -- they do seem to change every year
because I don't really care if I see the same one and I move around
-- she says, "Well, everything looks wonderful.
Your eyes are all clear.
Not a trace of retinopathy, very good."
Then she says, "You must be really good at following orders."
All I could think was, you really think I do this for you?
Man 7: So I just have a small point to make and maybe this is
closing the loop because I want to bring this all back to some of the
issues that Paul and I started with in terms of data.
But I was struck by some of the comments that have come out as
we've been discussing this about the implicit logical beliefs about
the value of steps down from 11.2 or whatever it was and how far
could you go and would it matter?
In some sense, I think I just want to raise our attention to the fact
that another source of data uncertainty is the translation of
biomarkers into risk and what that relationship looks like.
Because there is good psychological evidence that
particularly in cases like this where we have concrete goals,
people naturally translate things and assume that it's a linear
relationship. And if I'm only getting a quarter of the way
there then I'm only getting a quarter of the benefit.
That's not always the way the biology works.
That's not always the way the relationship works.
And so uncertainty about that relationship is at least as
important as in some sense what the number is.
And I'm not a clinician.
I don't want to claim to answer that but I think it's interesting
that even in this discussion we seem to have implicitly taken on
that logical linearity.
Whereas, if this is an exponential relationship that drop from eleven
to ten might in fact be a huge drop.
I don't know but that's something certainly to consider.
And it's a part of the discussion of what needs to happen in the
shared decision-making process between the patient and
the clinician. Paul Haidet: Well, the other thing
that I'll bring up about uncertainty. On the ***
side of the equation I'm worried that if I start shooting insulin I
might start using again. So what is the risk that he's
actual going to start using *** if we go on insulin.
So here we have a scenario where it's not just
uncertainty or certainty.
We have both uncertainty and certainty in the same equation.
And I find myself wondering is part of that pull as a clinician
that I feel by the glycohemoglobin because I'm running towards this
certain half of the equation here?
Liana Fraenkel: So it's interesting, I think,
the different reactions.
I'm not a primary care doctor and I assume for this discussion that
this was an even more than a monitoring event because this was
important-- the patient would derive important benefit from it.
So this is medically the way you'd want to go unless there
was a major problem.
It's obviously much grayer if that's not even an issue.
But I don't think truthfully I would go there if it wasn't a real
tangible benefit because the potential risk.
But it's interesting that from my perspective, I think,
because I've been there several times with patients.
It is usually in the veterans that's an unacceptable risk.
I don't even think that even the most paternalistic of physicians
at the VA or any place where there is a strong substance abuse cohort
would they be strongly recommending use of needles or
injections in a patient where there was really any tangible risk
of somebody restarting something as potentially lethal as abuse.
So it's interesting actually that there's even variability in how we
see our roles and that kind of discussion.
Paul Haidet: I'll give you the wrap up to the story.
When I'm in the room with patients my usual litmus test for good
decision-making is I like a decision or a plan that at the end
of the day both the patient and I are invested in the plan.
That we both feel some sense of responsibility for the plan.
And that we feel I'm good with this.
I'm good with this plan.
And so to me that's what I'm trying to work towards.
With this particular patient I was feeling the pull of the
glycohemoglobin and when this story came out I thought now
we're really between a rock and a hard place.
On the one hand there's this problem with diabetes.
And on the other hand, there's *** abuse.
And so we decided in that visit there was no insulin in that
visit. And actually it started an ongoing conversation over
the upcoming year, part of which was driven by my feeling
the compelling nature of the glycohemoglobin.
But it was a conversation about potentially see any way that he
could be taking insulin in such a way that he did not feel like
using. And so we explored a bunch of different kinds
of things. We had a visit where we had a whole conversation
about the difference between IV injection and subcu injection
but the problem is subcu injection feels like skin
popping. And we had a bunch of discussions about this about
what does this feel like and what will this be like.
In the end, he did end up on insulin.
My fear is that somebody came in and just said you're going on
insulin and here it is and forced it on him.
But as far as I know, he did not start using again and I like to
think that those conversations prepared him for the day when
somebody came and put the hammer down.
Paul Han: One quick comment.
Just to riff on what Brian was just saying and also Clarence,
I mean I do think what's interesting about this is case is
that it points to the use of a heuristic by physicians.
We tend to look at biomarkers -- cholesterol levels, PSA values,
and really forget about at what risks these symbolic
representations actually represent themselves.
An an exploration of what those risks actually are might inform a
discussion about goals and at least consciousness of goals
rather than considering these as things in and of themselves that
have kind of a life of their own.