Tip:
Highlight text to annotate it
X
Mira Irons: So I'd like to introduce myself. I'm Mira
Irons, I'm the chief of Clinical Genetics at Boston Children's Hospital, and I'm here
representing the APHMG, but -- and I keep getting moved up on the schedule. So I guess
this is a good time, I can -- I can really try to set the tone for this. I'll either
upset everybody or move us all in the right direction.
But my charge has been to talk about fellowship training in genetics and genomics. And just
to give you a sense of who I am, I'm kind of like Sibyl here. I come here as a medical
genetics educator. I ran the Harvard training programs for 10 years. And I have to say that
it is wonderful to see two of my former fellows at the table at meetings like this because
that's what it's all about, as a genetics educator.
I also come here as a clinician. Those of you that know me know that I am a clinician
at heart, and I really do feel that doctors learn things by taking care of patients.
They can go to all the meetings in this world, and they can read all the journal articles,
and look at all the cover stories in the world, but what they will remember is what they apply
to a patient.
And I also come here as a clinical chief who gets at least five calls a day from residents,
fellows, and people in practice saying, "I have this patient here and I'm really not
sure what to do about it, and, you know, there's this family history of hemochromatosis, and
should I send the family to you to do this testing, or can I do it from the office?
And if I do it from the office, what lab do I send it to? And how do I interpret this
result? And why is the lab telling me that I need informed consent? And I have only
six minutes to see this patient." So, as I see it, that's the problem.
So the vision that we all had is -- and I'm dating myself a little in this, talking about
this slide carousels --
[laughter]
But all the geneticists -- all the geneticists, medical geneticists in the world -- and I
started my training in medical genetics in 1983 -- had this slide. "Once we know what
the gene is and understand the monocular mechanism of the condition, we'll be able to treat it
effectively and improve health." And we walked around for 25 years with this slide in our
slide carousel, and we're finally here. I think we're at the point where this is reality.
But a lot of specialties really don't know that it's reality.
I had a talk -- I was at a meeting a few weeks ago, and I talked with two internists about
genetics and genomics. And I said, you know, what about training some of your fellows and
your -- and your residents in genetics and genomics, and the response was, well, you
know, in pediatrics and in OB that makes a difference, but I really can't see this being
important in internal medicine.
[laughter]
And I think that that's really where we have failed. I think we failed because we haven't
taught people how it can be important. So we're there, but I would say that we haven't
and aren't teaching people how to apply this to their practice. And I know that a lot of
people in this room are really doing a lot of work at the local level, teaching their
residents and fellows and students the principles of genetics, so I don't mean to say that this
isn't being done, but I'm taking the 35,000-foot view here and looking at the calls. I think
we're teaching people the principles of genetics, but we're not teaching people how to apply
those principles to their practice and their patients, and that's the charge here.
So, as I said, we're finally at the point where genetics and genomics will impact medical
care, but we haven't given health care providers the information they need to use this effectively
and practically. [inaudible] payors are actually going to call the question here.
Genetic testing is really expensive. And it's being used appropriately by some people,
but it's also being used inappropriately by other people. And so what is starting to happen
is that payors are starting not to pay for testing that actually really can help with
the care of their patients. So we're at this funny crossroads in life where we can
finally apply some of what we have learned but we're going to be told that we can't apply
it.
So how do we teach genetics? [inaudible] Well, we teach it in medical school, and I think
that all of the medical schools in the country have genetics integrated in their curriculum
somehow, either standalone courses, or it's integrated into other courses. And we do this
really well. But most of us teach this in year one. I mean, I spend 10 hours a week
for five weeks with the first year medical students at Harvard, and we go through four
or five really difficult cases that teach people how to practice genetics, how to apply
genetics and genomics to patient care, and then I see those students as third and fourth
year students on the wards and they give me -- you know, the question I hear is, what
color tube do I send chromosomes in? And you know, how do I send off genetic testing and
what tests should I send? Now, there are medical schools in the country, I know, that are bringing
this back, they're bringing the students back in the third and fourth years, but too many
aren't. Then you know, they graduate from medical school and they go into GME programs.
And how many program requirements contain any mention of genetics and genomics?
Bruce put together a meeting -- is it nine or 10 years ago? Right after you went to UAB.
Bruce Korf: 2004.
Mira Irons: 2004. Where he brought together some training
program directors in genetics, internal medicine, pediatrics, and family medicine, I believe,
for three days to sit there and talk about how to teach fellows and residents in the
primary care specialties in genetics. And the message that came through loud and clear
was, if it's not in the program requirements, we're not going teach it. We may not know
how to teach it, but we also don't know where to put it. And when I talk to people on my
local level about this, the response is, well what do you want me to take away? You know,
if I put genetics and genomics in, then I got to take cardiology away, or I got to take
nephrology away, or, you know, infectious disease is really important. So I'm here
to say that maybe this is the time to say genetics and genomics is important, and it
needs to be in all of the program requirements somehow.
And then we get to CME ,and now MOC is the new player on the street, and how do we teach
all of these doctors that practice, who didn't have the advantage of going to medical schools
and training programs where they were taught genetics and genomics, how to apply this when
they have six minutes to see a patient.
So this has been my strategy. This is the embedded -- my embedded journalist or embedded
geneticist strategy, and I didn't do this to teach, I'll be perfectly honest, I didn't
do this to teach residents and fellows genetics, but this was actually a nice byproduct of
this. I'm fortunate enough, at the hospital that I'm at, to have 15 geneticists and eight
genetic counselors. And what I was seeing was that we were sitting on the 10th floor
of the Fegan building, which is the clinical building at Children's, waiting for patients
to come to us. You know, people would send us consults, outpatient consults, or we would
go on the wards and we would see patients, and those were the only patients that I was
seeing.
And if you take cardiology, for example, there were 35 cardiologists all seeing patients.
Perhaps funneling those patients to 14 different geneticists who weren't able to see common
patterns; everybody was kind of doing a one-off. And so I convinced the chief of cardiology
to basically buy part of a geneticist. You know, my pitch to him was, geneticists are
cheap, you don't have to pay them what you pay a cardiologist, and if we put the geneticists
in your clinic, it will actually improve patient care; the patient can see the geneticist and
the cardiologist in the same space. It could facilitate research, and you may actually
be to be able to raise some money for doing this.
So he took the challenge. He actually raised five years of salary for the geneticist with
that. And a really interesting thing happened from it: patients loved it. Families loved
it because they didn't have to go to three different places in the hospital to see their
specialists, and some clinical research studies came of it. But, all of a sudden, the cardiology
fellows were asking that geneticist to come to her clinic. In a million years they never
would have asked to spend the day in my clinic up on Fegan 10, and even if they did, it wouldn't
have been a good thing for them because they maybe would have seen a child with autism,
or they might have seen a child with polycystic kidney disease, but what they really want
to see are kids with cardiac disease. And all of a sudden it was relevant to them because
they were learning genetics within their scope of practice. And we did the same thing with
plastic surgery, and we did the same thing with orthopedics, and we're doing the same
thing with otolaryngology, and all of a sudden it dawned on me that people actually have
to learn this within their scope of practice.
So what are the possible solutions -- and this is my last slide for the timekeepers
up there -- is we have to, I really believe that we have to teach genetics and genomics
within a health care provider's scope of practice. And I'm sorry to have to point to GME, but
that maybe the best place where this should be. What do I mean by, is this a procedure?
We are looking at our hospital of introducing next-generation sequencing into clinical practice,
and I think it's come to a time where we're going to have to credential people to actually
order this testing. Until we teach people how to use genetic testing appropriately,
we're not just going to be able to open the flood gates and allow everybody to order this
test. So we're looking at credentialing people to do this like you would to credential laser
privileges, or endoscopy privileges at the hospital, which is kind of a new twist; people
don't usually credential people for cognitive things.
From a GME standpoint, you know, perhaps I haven't -- I can tell you that -- looked at
other program requirements out there, but maybe we should look at the program requirements
to actually see what's there. From a CME standpoint, I think we really need to consider
point-of-care educational materials; you know, something that a health care provider can
find in 30 seconds in their office to help them take care of their patients.
There aren't enough medical geneticists in the world to see all these people, and we
shouldn't be seeing these people. One of my first year medical students at Harvard
this year, we were talking about genetic testing or something, and one of the students said,
"Oh, we'll just send them to the medical geneticist." And he turned to that other person and said,
"No, no, no. You should be able to explain this to your families." He's going into
family medicine in Michigan. And he said, "You know, I need to be able to tell people
this." You don't need a geneticist to do the easy stuff; you need a geneticist to do
the difficult things.
And then my last appeal is, use medical geneticists to work with all of your specialties to create
the educational content and provide consultation when needed. You know, there are a lot of
us -- there aren't a lot of us as a specialty, but we're in all your specialty organizations.
The American College of Medical Genetics and Genomics have done a lot of work on their
ACT sheets and their practice guidelines. The Association of Professors of Human and
Medical Genetics is a group that's dedicated to education. And I actually think that
if we have a coordinated effort to do some of this, we can make it happen.
So that's it. Thank you.
Marc Williams: Thank you, Mira.
Mira Irons: [inaudible]
Marc Williams: I'll just add one comment. I think the credentialing
idea is a very interesting one. I will note that some payors are doing this sort of de
facto in the course of utilization review, and that if a genetic test comes in, ordered
by a geneticist, that is approved, whereas if it's ordered by nongeneticist, then there's
a second level of review that takes place. And, actually, we'll have some questions,
so you might want to just at least stay close to a microphone if you can. So I think
that it's -- the credentialing idea is innovative but it -- and it's not without precedence.
So I think I've got --
David Beck: It's David Beck --
Marc Williams: -- David first, and then Bruce, and then Gail.
David Beck: So I'm David Beck. I'm at the Medical College
of Wisconsin. A bunch of people around here know that we're using whole genome sequencing
clinically in our hospital, and the hospital leadership, along with us, have -- are basically
going to -- or we're definitely planning to require credentialing for anyone who's going
to order a whole exome or whole genome, so that's underway.
Mira Irons: I think that at smaller institutions it may
be actually easier to put the brakes on these things, but, you know, for example, my institution
has 3,000 health care providers that can order tests. I have 900 rotating residents that
come in and out of that hospital every year. So, you know, that's the difficulty with some
of this is people come and order a test, and then they're gone. But I think credentialing
is really the best way to do it.
David Beck: Yeah. I mean, I think the issue here, obviously,
is that there are a few thousand, you know, attending physicians and so forth in the hospital
system, and, basically, when we've talked at sort of higher levels with individuals
about this possibility, there actually wasn't much pushback at all because most people feel
that this is a complex test, very expensive, and that they're -- and that a lot of physicians
don't feel super comfortable ordering it.
Marc Williams: Thank you. I'm going to interject just for
a second to people that are note taking, when I hear something that I think is something
that I want to put in the parking lot for later discussion. And so one of the comments
I would just make after David's comment is, we would -- if we go the credentialing route,
then we would need to think about criteria for training and competence; that would be
something to discuss. Bruce.
Bruce Korf: I just wanted to underscore a point that Mira
made. This embedded geneticist is not a cardiologist interested in genetics, it's
a geneticist interested in congenital heart disease. And it may not be obvious to everybody,
but genetics is ACGME accredited residency and an ABMS-recognized specialty. And I think
where we need to evolve is towards recognizing subspecialty tracks in genetics, realizing
that genetics is such a broad field at this point, it's all of medicine, and we have to
provide training opportunities for people who want to sub-specialize in areas like cardiology,
or oncology, or many others.
Marc Williams: Thank you, Bruce.
Gail.
Female Speaker: First I want to thank Mira for a great talk
and also all her work in education, and then briefly say, that Arno Motulsky, back in the
early 1950s and early '60s, used to find patients by walking around the wards and looking at
the charts. And in HIPAA era, unfortunately we can't do that, but we luckily have good
connections into internal medicine because of that. But then to reflect on something
you just said, half of medical genetics residency spots every year do not fill. They're empty,
we can't find trainees for those spots, and there's a lot of reasons why, and I don't
think we probably have time to go into it. But I think this is something that really
merits attention, and part of it is, as you say, it's the subspecialty. People want to
do cancer genetics or they want to do adult genetics and not learn dysmorphology. So I'm
not sure it's just sub-specializing after medical genetics, but possibly different tracks
where you don't have to learn everything to be in at the beginning.
Mira Irons: I think you're absolutely right. We have the
Medical Genetics Residency Program Directors Association that's been grappling with this
for years, and I think what happens is that medical students, in their first year, get
really excited about genetics and genomics. I mean, they really -- you can see in their
faces, they're really excited, and then they fall in love with an organ system. You know,
you go through this vine thing when you're a medical student: do I like kids or adults,
you know, that's your first thing; you know, am I going to be a surgeon or a cognitive
specialist; and then they fall in love with an organ system, and then to do genetics is
another two years. And I think what Bruce's saying, and what a lot of us have been talking
about, is it doesn't have to be an either/or anymore; there can be subspecialty tracks
in genetics that that can be user-friendly.
Female speaker: But to even take your boards, you have to
be very -- you know, I'm a internist, so I learned all that pediatrics and, you know,
biochemical [laughs] genetics, but not everyone wants to do that. And we find, in particular,
when we're bringing oncologist into our training program, they're happy to do a year and become
sort of experts, but there's no certification for that, there's no acknowledgment of what
they've done, because they can't sit for the boards unless they become a generalist.
Marc Williams: So I'm going to let Bruce just have one brief
comment since he has spent a lot of time in this area.
Bruce Korf: Yes, so the American College of Medical Genetics
and Genomics has a task force on the issues of genetics education, and this issue has
come up a lot as a topic for discussion: how can we provide training opportunities for
people who are specifically interested in cancer genetics, just to take an example,
that doesn't require them to learn to diagnosis congenital anomaly syndromes, maybe other
than ones that are associated with cancer. And this is very definitely an area we would
like to pursue and the American Board of Medical Genetics is open to, and we would be delighted
to have partners in other specialties to pursue it.
I will tell you one of the challenges is fairly expensive thing to do because you need to
create an exam, which, with the NBME, the price of entry to do an exam is high, so somehow
it's going to have to be jump started, but I think you'll find tremendous hope in this
-- to this.
Marc Williams: Thank you, Bruce.
Gene.
Eugene Passamani: I wonder, Dr. Irons, have you thought about
your model of embedding geneticists outside the walls.
Mira Irons: In private --
Eugene Passamani: Yeah.
Mira Irons: Private practice? You know, that would be
great. I think we don't have the manpower. You know, it's a manpower issue. And I think
we just don't have the manpower to do that.
Eugene Passamani: But it gets, ultimately, to teaching --
Mira Irons: Right.
Eugene Passamani: Teaching where it hurts --
Mira Irons: Oh, you're absolutely right. You're absolutely
right.
Marc Williams: I'll just note that at Northern California
Kaiser, they are actually experimenting with embedding genetic counselors within primary
care pods, much the same way of having them available to enhance family histories and
discuss that, and as a place to be able to, again, do some education, do some consciousness
raising, and also to serve a triage function. I think they found that that's been reasonably
efficient. Now, again, the Kaiser model, at least in terms of the economics of the
delivery, are not generalizible, but I think it's something interesting to think about.
The other thing to put on the table with relation to that is whether or not we need another
level of genetic provider. That would be one step below the training of genetic counselor;
you might think about it as a genetic educator, something of that, an individual who could
take a family history, who could analyze a pedigree, and could also serve a triage function
that would be less resource intensive than geneticist/genetic counselors, and whether
that is a gap that needs to be filled.
Male Speaker: A barefoot doctor, in other words.
Marc William: In a way, yeah. Whatever that is, sure. Mike.
Michael Murray: So I was just going say that as far as a relevant
example, that credentialing communities can understand for intellectual input into interpreted
services, we can always point to EKGs is something that they all understand. And there's only
certain doctors in the hospital that can officially interpret them.
Marc Williams: Thank you, Mike.
Pearl.
Pearl O'Rourke: I really like the idea of credentialing for
ordering, but I think maybe to put in Marc's parking lot is a fact that a lot of this is
being done in the research environment, and there's no credential for being a researcher,
let alone for ordering a test, and I think with the blurring of those boundaries, we'd
really need to look at the ripple effect.
Marc Williams: Yeah, so, I think that's a good point. And
Pearl, as you well know, there's a lot of discussion in a lot of different venues about
the idea of the blurring between the clinic and research, and so I think that while that's
clearly an important topic, it's something that this group is cognizant of, and I think,
actually, we spent some time on this at GM II. I wouldn't necessarily want that to
be the major focus of this meeting, but certainly always good to keep that in mind, and you're
absolutely right. In a research space, the concept of credentialing is not relevant.
Mira Irons: But actually, Pearl, just brought up a -- it
made me think of a great thing. You know, most of us in the research environment have
to take some online, whether it's the city course or something like that, to be able
to participate in research, and that may be actually kind of an easy way of solving this
at the national level, is to create some sort of an online curriculum for ordering genetic
tests so that every hospital doesn't have to come up with their own educational curriculum
and credentialing. It may actually solve both problems.
Male Speaker: The one question I have, with doing specialist
view would be, if you are giving whole exome or whole genomes sequencing back, I mean,
you're not just talking about pediatric disease, you're not just talking about cardiac disease,
you're talking about everything. And can that specialist then talk about everything, or
is that going to end up being still a geneticist or someone else?
Mira Irons: Well I think that -- you know, I think specialists
can actually talk about the results within their own specialty, but maybe that's when
you use the medical geneticist for the other stuff. You know, for the secondary findings
that you may find that aren't within your specialty, but at least teach the specialist
how -- because I'll argue that the nephrologists, or cardiologist, or the immunologist knows
those genes better than I do, you know, and so they should be able to interpret those
results for their patients, but the secondary findings they need -- may need some help
with. And that's where maybe genetics should come in, is to help with that.
Marc Williams: Debra.
Debra Leonard: So in -- hi, Mira. I'm over here on the -- to
your left. Hi. In pathology, it is routine to have certain tests only orderable by someone
that we trust ordering that test, and you could -- that's either down through the EHR
that when you order the test, there's a block saying you need to get a consult or, you know,
from someone. So the same thing could be done with genetic tests, and it's a little
different then the credentialing process. But you may want to consider both pathways.
Mira Irons: I think it depends on the size of your institution.
You know, when you're having thousands and thousands of outpatients a day, you can't
call a consult for everything. And I think some of the larger hospitals have been doing
that. I know that -- I think wasn't it at the Brigham, Mike, that every time a certain
test, the genetics fellow got paged, and, you know, that's just not reasonable. You
know, you can't -- the genetics fellow really should not be the triage person for telling
full professors whether or not they should or should not order tests. We have to teach
people how to do this. We can't put the block at the consult level.
Debra Leonard: I agree on the educational process, but in
the meantime, it may be a stop-gap measure.
Marc Williams: Yeah, and I think that there's actual literature
that the area ARUP White Paper that came out that looked at having genetic counselors contact
the ordering physician for all the genetic tests that identified an average amount of
$36,000 of testing a month that was inappropriately ordered, is clearly an indicator that not
only are there patient quality issues related to inappropriate testing, but there's dollar
amounts that are with this as well. So -- and there's also a couple of papers that have
come out in the oncology world that has also showed significant issues with inappropriate
testing. So I think we have empirical evidence that shows that this is a significant issue
and that it does impact patient care and cost, and so solutions at the practice level are
definitely needed.
Geoff.
Geoff Ginsburg: Thanks. Geoff Ginsburg. I just wanted to raise
the point about genome interpretation, and I think it will be really interesting to see
how some of the commercial firms that are evolving to do genome interpretive services
will, but this conversation about how we, as a professional -- how professional societies
are thinking about genome interpretation. So there's commercial -- there's a commercial
strategy that's going on right now to do whole genome interpretation and sell that to paying
customers, and we're talking about trying to create standards and training to do this,
you know, and these two are happening in two different worlds.
Mira Irons: Well, they're going to issue a report, and
then they're going to hope that whoever ordered that test and gets that report knows what
to say to the patient. And that's the rub; that's where it stops. Because I think that
some people might know how to interpret that report for the patients, but others may not,
and it's -- you don't know what you don't know until you get there. And I think that's
my concern.
Marc Williams: Bill.
William Zoghbi: Talking about appropriateness of testing,
the American College of Cardiology over the past seven years or so have come up with appropriateness
use criteria for different scenarios; for noninvasive testing, for [unintelligible],
and other things, and I think this is this is the good idea to embed appropriateness
of genetic testing depending on the clinical situation.
Marc Williams: Thank you. I think we're at the end of this
discussion. This was really great. I heard three take-aways. One related to the credentialing,
one related to the idea that we need to consider different types of tracks of genetics training,
and then something that we didn't really spend a lot of discussion time on but which is personal
interest of mine and some others in the room, is the idea of point-of-care information and
education. So we'll just park those. And next we have a couple of quick hitters from the
MEs, and we'll start with Dr. Nasca from Accreditation Council for Graduate Medical Education.