Chapter 18: Creating MD Anderson’s Practice Algorithms; On Blending Art and Science in Medical Practice: Practice Algorithms and Targeted Therapy
Files
Loading...
Description
Dr. Rodriguez tells the story of MD Anderson’s 147 Practice Algorithms beginning with the origin of this initiative in the 1990s movement to define “pathways of care.” She talks about the process of establishing an algorithm and discusses the effects. She also notes the different reactions of clinicians, who may immediately adopt the algorithm or who may take convincing. Dr. Rodriguez talks about the dangers of dogmatism in medicine. She notes that medicine is both an art and a science, but the poles need to be harmonized in order to be humane. Dr. Rodriguez notes that limits of targeted therapy and sketches an emerging view that this approach will be replaced by a focus on failures in the body’s surveillance and regulation mechanisms. She notes committees in place to support clinicians as they self-monitor the quality of their practice.
Identifier
RodriguezA_03_20150501_C18
Publication Date
5-1-2015
Publisher
The Making Cancer History® Voices Oral History Collection, The University of Texas MD Anderson Cancer Center
City
Houston, Texas
Interview Session
Topics Covered
Building the Institution; Building/Transforming the Institution; Multi-disciplinary Approaches; Growth and/or Change; Understanding the Institution; Professional Practice; The Professional at Work
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
Disciplines
History of Science, Technology, and Medicine | Oncology | Oral History
Transcript
Tacey A. Rosolowski, PhD:
Yeah. Interesting. Other areas of function within Medical Affairs—
Alma Rodriguez, MD:
Well, you mentioned the issue of the algorithm. So of course, you know, I said it’s really important that we have the individuals with the right credentials and the appropriate competence to perform the job. We also want to provide them with the right tools to perform their job.
Tacey A. Rosolowski, PhD:
Hmm, OK.
Alma Rodriguez, MD:
And so it became apparent, even predating my coming into this role, somewhere in the 1990s, there was a movement nationally as well to establish what were called “pathways of care,” and this was particularly true in surgery where, again, organizations, just HMOs [Health Maintenance Organizations], and so on, were pushing for the delivery of care within X-number of days, within X-number of hours, and you know, the whole trend to efficiency in how patients were moved through, if you will, moved through the system of the hospital or the clinics.
Tacey A. Rosolowski, PhD:
And I guess standard it too—
Alma Rodriguez, MD:
Standardizing it—
Tacey A. Rosolowski, PhD:
So you realize what you’re paying for.
Alma Rodriguez, MD:
Exactly. Exactly.
Tacey A. Rosolowski, PhD:
Yeah, OK.
Alma Rodriguez, MD:
So even back then, there was some movement to start to begin to look at the processes of delivery within the organization. Nothing much happened out of that, other than some groups did map out what their care processes were. But eventually, when I was assigned to this role, I realized that on a national scale, we also were beginning to talk about algorithms of care, actually they were called “guidelines,” guidelines of care. And there’s a whole debate around the terminology, what is a true guideline, what is a pathway? We chose to call our maps of care “algorithms,” because essentially, it was, like, if this, then that. If that, then this. You know, so it gave essentially a map. Essentially we mapped out processes of care. And within those maps of care, there were unique focus areas that we felt needed a deep dive in, particularly all the domains that had to do with the delivery of chemotherapy.
Tacey A. Rosolowski, PhD:
Can you give me an example of what one of these algorithms might look like?
Alma Rodriguez, MD:
Oh, I can show you on our Website—
Tacey A. Rosolowski, PhD:
Oh, sure.
Alma Rodriguez, MD:
If you want to see them.
Tacey A. Rosolowski, PhD:
And then actually, if you could, then maybe I could ask—I’ll remind you to maybe send me a shot of it so that we could—
Alma Rodriguez, MD:
So when will—
Tacey A. Rosolowski, PhD:
And ooh, so we’re worried about our recorder, here—
Alma Rodriguez, MD:
Well, I think this one has a—or some of these [inaudible]. So let me show you here in this book, textbook that we published on cancer survivorship— We developed algorithms for survivorship care, of course. So today, we have algorithms of care for several domains of care; for cancer treatment, for survivorship, for prevention and for what we call “supportive care,” or, “medical supportive care,” I forget what the subheading is. But it’s about managing other associated problems, such as preventing thrombosis, prophylaxis for deep vein thrombosis, management of pneumonias, whether they are related to community infections or hospital-acquiring infections. Management of chest pain and myocardial infarction, and so on, so that we are taking into account the more common complications we see, as well as the actual treatment of the cancer itself. So this is what they look like. I mean, essentially, they’re a map. And the map says, “If this, then you must do that.”
Tacey A. Rosolowski, PhD:
OK, so myeloma post-treatment and NED, which means--?
Alma Rodriguez, MD:
No evidence of disease.
Tacey A. Rosolowski, PhD:
Oh, OK. Yeah. So then you go through surveillance, oh I see. And then it’s, like, selecting from a flowchart.
Alma Rodriguez, MD:
Exactly. Exactly. And so every map--
Tacey A. Rosolowski, PhD:
Wow, one year of age and up, four years of age and up—
Alma Rodriguez, MD:
But this is [inaudible].
Tacey A. Rosolowski, PhD:
[inaudible] diagnosis. OK. Got you. OK. Interesting.
Alma Rodriguez, MD:
So many—so—
Tacey A. Rosolowski, PhD:
I can’t even imagine the database that you would require to put together something like that.
Alma Rodriguez, MD:
So we have that. (laughs)
Tacey A. Rosolowski, PhD:
Yeah. Yeah.
Alma Rodriguez, MD:
And we established some ground rules, so we say, for example, in survivorship, all your algorithms must contain four domains. One domain is the surveillance one, but the other one is also monitoring for late effects, early detection and risk reduction and psychosocial functioning.
Tacey A. Rosolowski, PhD:
OK.
Alma Rodriguez, MD:
You must address these four domains; tell us what you would do for your patients in these four domains.
Tacey A. Rosolowski, PhD:
Right. OK. Did you work with Lewis Foxhall on this? I think he mentioned to me something about he uses different domains—
Alma Rodriguez, MD:
Well, he was a co-editor of this book.
Tacey A. Rosolowski, PhD:
Oh, yeah, there’s his name. Yeah.
Alma Rodriguez, MD:
But the domains of survivorship, we worked with all specialties. Each specialty really designs its own survivorship domains. We took those four domains from the Institute of Medicine report on what constitutes good survivor care. So we said, OK, there are national estab—or national recommendations and what should be the domains of care that survivors receive, so let’s be sure we built our care—models built on those domains.
Tacey A. Rosolowski, PhD:
So I can see the advantage here. Now, I guess I was making the assumption that these practice algorithms were created diving into MD Anderson databases on treating thousands of patients with all sorts of different cancers. Is that the case?
Alma Rodriguez, MD:
No. You’re talking about, then, outcomes analysis.
Tacey A. Rosolowski, PhD:
OK. OK.
Alma Rodriguez, MD:
You’re talking about analysis. This is about simply creating an informational work tool for physicians.
Tacey A. Rosolowski, PhD:
OK.
Alma Rodriguez, MD:
Within those informational work tools, then, there will be areas where you have to do a specific tool for that performance. So again, so if you say for breast cancer, we just updated one, for example, for invasive breast cancer, limited stage. So the recommendation is you must do either doxorubicin or Taxol-based chemotherapy if the patients are hormone receptor negative and they don’t have Herceptin. OK. So for that subset of patients, then you do Taxol and doxorubicin chemotherapy, which regimens, so we create order sets for those regimens that map—then, along with that algorithm.
Tacey A. Rosolowski, PhD:
OK, interesting.
Alma Rodriguez, MD:
So anyhow, so this is a theory—this is part of how one then develops also quality controls and quality measures, because we can then say one of our quality measures can be, do you provide—in fact, this is one national quality measure for patients who have estrogen receptor, progesterone receptor positive breast cancer, do you give them hormonal therapy, which our algorithms say you should. So do you walk the talk of your [inaudible]?
Tacey A. Rosolowski, PhD:
Interesting!
Alma Rodriguez, MD:
So, no, I think you’re talking about measuring outcomes.
Tacey A. Rosolowski, PhD:
OK.
Alma Rodriguez, MD:
Looking at outcomes. That’s the Tumor Registry. And we have a Tumor Registry, the Tumor Registry follows patients for periods, you know, for their lives to find out how they are doing, and have they relapsed, and are they still alive? So, for example, these are the survivors for cervical cancer, early stage, across many decades. You know, you can see that. And what is interesting is that for some cancers, the outcome has always been good, even before—even in the 1940s. And we haven’t made much difference, and for some, we’ve made a huge difference, and for others we have made absolutely no difference, and the outcome is really still horrible, despite seventy years.
Tacey A. Rosolowski, PhD:
Wow, amazing! Amazing.
Alma Rodriguez, MD:
Yeah.
Tacey A. Rosolowski, PhD:
Now, in putting together these practice algorithms, how did that happen?
Alma Rodriguez, MD:
So that, by the way, is run by the Department of Clinical Effectiveness.
Tacey A. Rosolowski, PhD:
OK.
Alma Rodriguez, MD:
They report to me also.
Tacey A. Rosolowski, PhD:
And when was that department established? Is that, did that—
Alma Rodriguez, MD:
Again, it sort of preceded my coming on board.
Tacey A. Rosolowski, PhD:
OK. OK.
Alma Rodriguez, MD:
Because they—once upon a time, they were supposed to be working and developing those pathways. When I took over the office, we tightened up the process. We said we will have institution-wide clinical algorithms, we will have for all the major disease categories, all the major cancers that we see, we are going—you can’t have an algorithm for everything; there are some malignancies that are so rare that you can’t, you know, really—there is no standard, if you will, or no known best strategy for treating them. But for all the more common and more widely-seen malignancies, we have developed algorithms for cancer care. We have over 100 now, 147 algorithms. And—
Tacey A. Rosolowski, PhD:
What’s been the effect of the algorithms?
Alma Rodriguez, MD:
Well, what has been the effect of the algorithms is in—number one, it brings to awareness, to people’s awareness, that there are indeed best practice processes. In a way, it’s an intellectual discipline process, it’s a process of doing a very rational and thoughtful analysis of where should we be? Then it usually piques the interest of people in saying, well, where are we? So some departments have, I would say, the most—the best outcomes have been that some departments have become interested in looking at themselves again, a self-inquiry, looking at ourselves and saying, “Gee, are we really doing this?” And, “Is this what we want to keep doing?” Some of them have questioned, well, you know, just because we have done X, Y or Z forever, it doesn’t mean that it’s the best strategy. What does the data say? And this is where Stephanie Fulton and her group come in, because they support us in doing fairly—very professional intensive literature searches. And they then can give us the objective information that says, well, you know, that’s changed. Other people think that this is better, or they might say, you know, the needle hasn’t moved, it’s still the same, in which case, it might also initiate a different conversation, which is, “Gee, should we start to try something different?” (laughter)
Tacey A. Rosolowski, PhD:
Yeah.
Alma Rodriguez, MD:
So, you know, so that’s been, again, in the best case scenario, it is, and for some departments, this has been a process of self-inquiry, of self-assessment, of updating, renewing, refreshing information on what’s appropriate and relevant to their practice. In others, in many of the supportive care algorithms, for example, the management of deep vein thrombosis, it has initiated major conversations about who are the appropriate patients who should be placed on these prophylaxis modalities of treatment, are we doing it? It generated a whole deep analysis into practices by various groups. And we sort of surfaced, who are the people who really do it, the people who don’t do it, and we fed that data back to them. And they’re like, “Ooh, that’s us, we can’t believe it!” So it again has brought—one department that said, you know, our patients are really high-risk, we understand. And yet we’re seeing this as a complication often. Why is that? So they initiated a research protocol for that. So it can have very positive consequences, depending on the attitude of the individuals who are participating in the process. And again, we don’t expect humans to be uniform. So it’s a good thing. I mean, it has generated a lot of very good things out of the process.
Tacey A. Rosolowski, PhD:
Was there anything in particular you learned from going through this process of working with all of these individuals, and taking this perspective?
Alma Rodriguez, MD:
Well, so I’ve learned that there’s some—well, first of all, I’ve learned that the overwhelming majority of the physicians who practice here care deeply about doing the right thing, and taking the best care of their patients. I mean, that’s been incredibly rewarding for me, to say as a profession, I think we are an outstanding group of people. I’m very proud to work with them and for them, actually, because I work for them. And so that’s been one thing that I’ve learned. The other thing that I’ve learned is that there is always a potential risk in medicine. I mean, this has been true for centuries; we are a profession that is very, for lack of a better word, dogmatic, and that you have to be vigilant to the risk of being purely dogmatic versus quality and safety-motivated, when you say, no, look, this is the best way to do this. It’s not OK to do X, Y or Z, just because you like to do things that way, right? I mean, there is always the—the physician is the artist. I mean, everyone says that medicine is both an art and a science. Well, one has to guard a bit against the over-artistic aspects, as well as the over-scientific aspects, because being at both extremes may not be the most optimal medium for the patient. I think that there is a certain—there’s a harmony to both the art and the science. I think that while the—and patients are very conscious of this, they truly do want the treatment that, according to the scientific evidence is the best, or would be the best. But at the same time, they want the treatment that would be most suited to them as individuals. So there is a—that’s the—I think that’s the most valuable professional skill to have, to have the appropriate judgment to determine the harmony of the science versus the humanity of the decision. If this is the right treatment, is it the best treatment for this individual?
Tacey A. Rosolowski, PhD:
Just the way you phrase that kind of brought to mind the whole issue of targeted therapy. And I’m wondering how targeted therapy fits into the practice algorithms at this point.
Alma Rodriguez, MD:
Well, there are some scientifically-proven targeted therapy strategies. And, in fact, the most general of the targeted therapy principles is that you decide the treatment according to the tumor type; you know, a breast cancer may not necessarily be the same as a colon cancer, as an ovarian cancer, as a brain cancer, and so on. Even within breast cancer now, we know that there are different types of breast cancer. So the algorithms are supposed to align to each of the sub-categories of malignancies. Now, having said that, there is now an expectation that we do the [inaudible], even to the molecular level of the individual. The problem is that only a handful of the specific tumor markers today can be meaningfully addressed, OK? And so to do a whole genome analysis of every single individual is, in fact, meaningless, unless you know that there are useful therapeutics for certain targets.
Tacey A. Rosolowski, PhD:
Right.
Alma Rodriguez, MD:
And at this point in time, we do have some scientifically-confirmed useful treatments for targets. But, I recently heard a talk from a world-renowned scientist, talk about this issue. He said, you know, even within a single tumor, we are finding that there is huge heterogeneity in the complexity of the gene. You know, so some cells will have this pattern, but other cells will have this pattern, and yet other cells will have this pattern. So which one of the malignant cells are you going to aim your tumor at?
Tacey A. Rosolowski, PhD:
Right.
Alma Rodriguez, MD:
So in fact, we’re shifting now our thinking, well, we haven’t abandoned the concept of targeted therapy. And, in fact, it’s a valid concept for many tumor types, for some tumor types. So for example in breast, we know that the presence of the HER2-neu receptor calls for treatment with the antibody Herceptin, because that helps many of the patients with that marker. It doesn’t help everyone who has that marker, OK, but it is proven to be of benefit to a large proportion of those patients. So therefore, we would use that treatment for that particular tracer marker. But at the same time, you know, what about all those people that didn’t respond to the targeted therapy? So we’re shifting now to a concept of thinking, how does the body fail to recognize that these cells are not normal, or they’re not healthy? Because our body has self-regulatory mechanisms by which, you know, cells that are damaged do autolyze; they kill themselves. So why is it that this control of self-regulation failed? Why did the immune surveillance fail to recognize the cells as essentially rogue individual cells that are doing their own thing, outside of the, if you will, the overarching domain of health control that the body has. So the concepts now are shifting towards, let’s look at the regulations for—let’s look at the mechanisms that have failed in the so-called immune surveillance of tumors, and that’s where all of the novel immune therapies that you probably have been hearing about are now coming to the fore, because that model of care is very attractive, in that it’s generalizable. You don’t have to have a specific gene marker; you simply have to have a recognition that that cell is not normal. And the recognition mechanisms are very generic, they’re broad. They’re not to a specific gene, or more specific DNA marker, or a specific—they simply are to recognizing this cell is not acting normal. And so that’s really the shift in paradigm that we probably are going to be seeing prominently evolve. It’s already becoming prominently recognized. So I think there’s a lot going on right now.
Tacey A. Rosolowski, PhD:
Yeah.
Alma Rodriguez, MD:
Yes, indeed, for some tumors, targeting a geno or genetic marker is appropriate. But those tumors are rare, the ones that are really sensitive to the one targeted therapy are rare. So more to come.
Tacey A. Rosolowski, PhD:
Absolutely. Absolutely. Is there more that you wanted to say about the practice algorithms, kind of next steps, or—?
Alma Rodriguez, MD:
Well, in the next iteration of the algorithms, we hope to be able to have more robust capabilities to self-monitor the practices; in other words, to say if you said that this is the best strategy, are you, you know, are you following what you said was the best strategy?
Tacey A. Rosolowski, PhD:
Is there any oversight right now of that?
Alma Rodriguez, MD:
We have the Medical Practice Quality Committee.
Tacey A. Rosolowski, PhD:
OK.
Alma Rodriguez, MD:
That was just named in November—
Tacey A. Rosolowski, PhD:
Oh, wow!
Alma Rodriguez, MD:
(laughs) —of 2014 by the Executive Committee of the Medical Staff. And these are the focus areas that we launched as critical for oversight of this committee are, number one, the access of patients to the Institution, the diagnosis, are we doing the right diagnosis in a timely and efficient fashion, and appropriate fashion? Treatments, treatment decisions based on best evidence and eliminating adverse outcomes for the treatments. So, safety guidelines as well as effectiveness guidelines. And then at end-of-life care, are we having the discussions in a timely fashion, and are we referring patients to appropriate support services? So essentially, you know, it starts from the beginning to the end of our cancer care domains. And we’re looking at the two very—you know, the start and the end, of course, are critical for the patient for their experience perspective. But these, the diagnosis and the treatment are very driven by expertise and appropriateness of care. So these two are huge, huge domains that we’re going to focus on from a medical practice perspective.
Recommended Citation
Rodriguez, Alma MD and Rosolowski, Tacey A. PhD, "Chapter 18: Creating MD Anderson’s Practice Algorithms; On Blending Art and Science in Medical Practice: Practice Algorithms and Targeted Therapy" (2015). Interview Chapters. 349.
https://openworks.mdanderson.org/mchv_interviewchapters/349
Conditions Governing Access
Open
