 I mean, thanks to the speaker so far and Dan, do you want to start for or okay? You're reserving the right for rebuttal All right, thanks very much those were really really helpful and instructive talks. I really appreciated them for me as a provider of medical genetic services the biggest obstacle I face is insurance coverage and so I think these Economic issues are key to our being successful We do I think need to find out from the payers What is the evidence that they want and hopefully if they tell us what evidence they want then when we generate it They won't ignore it. I think that Julie was right when she said we need larger data sets. We need something that's going to be convincing to them I think the scale of the projects are an issue here We also probably need to sell ourselves a little better not just to the insurers but to our health systems because Genetic medicine is a very small field in medicine and it's very easy to get lost in cardiology and oncology and diabetes care and You know if we are trying to position ourselves as leaders in health care We need to be in at the table and we're not Often and so when our health system makes contracts with insurers They're worried about reimbursement for caths. They're not worried about reimbursement for genetic tests And the result is however, they don't want us ordering things that they're not going to get reimbursed for so it's a problem So we need to be represented there We did a study led by David Veenstra who I think my contribution to economics might be just to clone David So expect that Grand Terry With and the first author was a talented young guy named Carlos Gago and we were looking at the right-sized panel for for Lynch syndrome colorectal cancer testing and We looked at different various sizes of panels and it was a nice solid and interesting result but the thing that interested me most about it is When they perform the sensitivity analysis, which is when you move a bunch of variables and see what happens the primary driver of Cost and cost effectiveness in particular to drive those qualities down was how many family members? you could follow-up test and what the benefit to those family members were and That got us very interested. We obviously were medical geneticists were interested in families But it got us interested in how do you motivate? Cascade testing in families, but I can tell you this is not something that insurers care about insurers Generally don't have the adult relatives of the patients in their system some insurance systems do but are certainly doesn't So that and as Bob, I think mentioned Medicare simply does not pay for testing of unaffected individuals for cancer For example, so how do we change our system to think about? Public health as a common good and not just this is the patient I have in front of me and this is how it's going to change this patient's management and You know our health insurers have been pretty frank in one case just saying most of these people won't even be covered by us In two to three years. We're not looking long-term much less at their family We're just looking in the very short term So I think that for medicine to be cost effective We really have to change that outlook that it's just saving money this year and now that matters to us I don't know how to move that but I think it's a bigger problem than genetics than we have I Think for medical genetics to be cost effective We need you know several things one is The data to show because there are tests that I certainly know will change the patient's management But yet I still can't get them covered or it takes me a huge amount of effort To get it covered taking a lot of money and resources and I often do turf that work to the to the testing company So so that's an issue. You know who's gonna pay remains a problem and Often honestly We're doing free testing for patients. We're lucky to be an academic center But we just have patients who need genetic tests who we know their genetic tests will be informative who no one will cover them So we just do them anyway. That's not really how healthcare should be delivered It's not going to be cost effective until the providers know what to do with the information so we had patient who had an exome and a research study he had Warfarin variants that Predicted sensitivity to warfarin. He was stable on a warfarin dose It was explained to him that this did not change anything with his warfarin It had more to do with what his dose was but he was on a good dose. Don't don't worry about it We got a letter from him a month later an email Really grateful for now he has personalized medicine because now his primary care provider Has found out that he's sensitive to warfarin and so took him off warfarin and put him on a different drug Even though the patient was stable and doing fine on warfarin now We asked to talk to the provider so he could explain to us what we don't know that makes that decision makes sense So I'm gonna give him an out that maybe there was something I don't know but he didn't want to talk to us So I don't know what that would be But this is not cost effective medicine this patient was put on a much more expensive drug from a drug that made Perfect sense for the patient because the provider didn't understand what the information met patients very happy good outcome there So, but I don't know what his co-pays look like So we really need the first 30 days That's right So so you know it's interesting So I mentioned the model of we'll have the genetic counselors train the physicians But I don't work in a health care system where we hire people to educate physicians We send physicians off to you know Different courses I guess but it in in research we train providers to do certain things But in the health care system you expect to be hiring doctors who know how to take care of patients and not expect to be training them So we need to get in at the early end of training physicians So they know what to do with this information or it's not gonna work cost effectively Fundamentally right now what I see is a small minority of patients who get excellent care I see patients in boutique practices who get genomic tests that they self pay for and by the way is implemented into their epic EHR You know by a company That they're also paying for But it's not translating to my patients because no one will pay for my patients And so I I really think this is the obstacle that we need to look at Covering these tests. These are the tests. We know work now much less the tests that we can imagine for the future so I told Ebony and and Heather that that I know nothing about economics and And they put me on this panel anyway I will I will tell you that that my my contribution to economic analysis about comes was Was a conversation I had with Josh Peterson about What five years ago Josh something like that where I said I was sitting in my office and we were sort of drawing and I and I said You know You should make a mark-off model of what happens to all the thing all the all the possible outcomes that that people Who get exposed to clop integral with normal or variant genetics have and then you can sort of sit there with an Excel spreadsheet and tweak the model. I think it would take you an afternoon so So so, you know four years at a u-grant later the model is almost done And so so that's my introduction to economics It seems it strikes me though as as I listen to this To the points that have been made over the last 45 minutes or so that that economics really is the sort of final common pathway to to a lot of the issues that we're struggling with in terms of in terms of implementation so The issue in no particular order the issues are that that we need good input data in order to do the right output Analysis so if we don't have good local costs if we don't have a good sense of what the costs are And I'm not even sure we we capture that because there are it's it's it reminds me a little bit of what people When people ask me what happens? How do you fund your biobank and I give them a number because the biobank has a budget every year and then I say well But that doesn't count the electronic health record maintenance part And so it becomes almost impossible to sort of cost to count an effort like that And it seems to me that we're in the same position Here although we have to make some some hard decisions the discussion that that gale just introduced about the Education of providers I can give you you know exactly the same story about people out there in the community getting a genetic test result and Acting on it in a completely in an inappropriate way My example is a patient who gets a defibrillator because they have a V us that turns out to be benign in an iron channel so so and and and So we have an educational mission that that falls that and that muddies the economic analysis So when you have people doing stuff like that Using genetic data you you it becomes almost impossible to get a Decent cost accounting the thing that resonated with me I have lots of little notes, but I don't think I want to I don't think I want to go through them The things that resonated there are a couple of things one is that this business of big numbers that Julie touched on We have to We have to have big numbers because because what we're interested in in this room is the small sets that come out of Those big numbers because it's the small sets that have that derive I think the greatest benefit from the kinds of testing we're doing and for each large experiment There's a small set that has benefit and if we do many many experiments. There'll be many small sets that will derive benefit The I write wrote down here genetic exceptionalism that this this discussion always reminds me of so Why are we agonizing over this over genetics when there's a whole bunch of other kind of testing that people are doing? from MRIs of the knees to other examples where There's there is no benefit, but or there's no demonstrated benefit people adopt that in practice anyway and I and I guess the the thing that resonated with me almost more than anything else was something that Bob said right at the beginning and Or pointed out that it's this business of of who covers the genetic testing Leave I'm talking about the germline now. I'm not talking about your your genetic testing. So so it's always the When we're whenever we do genetic testing at least in my clinic We we always do it with with an eye on the family The diagnosis is made in the patient who's sitting in front of us And we don't need the genetic test to decide what we should be doing with that patient what we need the genetic testing for is Is to manage the rest of the family So so if there is a mandate that NHGRI can adopt or people in this room can adopt That that will help drive the field forward. It would be sort of Gina 2 I know that you guys can't advocate your kind of lobby for that stuff But it would be something along the lines of Gina 2 that that allows us to figure out a way to get payers to pay for a test that doesn't necessarily Benefit the patient in front of them or the patient that they're taking care of but allows us to sort of improve the general health of the of the families that that that they're part of so I don't know how to do that, but it seems to me that that's an innate. That's a genomic medicine Mandate so I think that those are the points that I wanted to make and I'll stop there Thank you Dan. Thanks Gail Open it up to questions. Noes Tony Yeah Tony Poland University of Maryland Dan I just wanted to add to what you just said and this is something that also came up at the payer meeting that I don't think I emphasized today, but Yeah, that that that the part about the family studies is actually is a huge piece of it obviously because yeah many of the things that we talked about So you talked about cystic fibrosis. There's also, you know, the cardiomyopathies where your your biggest benefit is primary prevention in Unaffected individuals and it really goes beyond that because there are many benefits to the patients and their families that don't accrue to the insurance provider So one thing that was talked about at the meeting and especially plus the exome sequencing is the educational system So so patients need children need a diagnosis of a particular condition for their IEP and That's not something that that's not a benefit that accrues to the insurance provider So how do you how do you make that part of it? How do you as you said make this about the or as Mark said about the public's health? So I so I don't know the answer that I don't know if anybody has an answer to how you you really you know We can talk about this being a problem forever, but you know, what are the answers for that? I Want to do a little experiment here By show of hands how many people believe that prevention saves healthcare costs no now let me I'll be for the question So let me tell you the results of an interesting story from the Netherlands The Netherlands looked at incurred healthcare costs and Dutch males over the age of 21 And they stratified it in three groups and what they found was that an normal weight non-smoking Dutch male incurred lifetime healthcare costs after the age of 21 of 250,000 euro an overweight Non-smoking Dutch male incurred healthcare costs of a 210,000 euro and a smoking overweight Dutch male incurred healthcare costs of 180,000 euro anybody want to venture a guess as to why that is They die right so here's a dirty little secret of healthcare economics dead people don't incur healthcare costs so we need to be very Circumstant about tossing out the idea that you know this saves money genomic saves money or prevention saves money Now from a societal perspective The fact that people live longer and are productive longer is a very good thing But the healthcare sector gets zero credit for that. We only get the cost and so I Only say that as a cautionary tale to make sure that we don't fall into the trap of You know looking at economics from the perspective of if we avoid this or if we if we avoid a sudden cardiac death That that's you know great for healthcare costs. It's actually a terrible outcome for healthcare costs for society It's a great outcome for the family. It's a great outcome So we just need to think about that a little bit more Globally and make sure that we're not Again raising the hype of what it is genomics can do when in reality our ability to move the the cost care Needle may be relatively limited But Sort of a quick comment with with Medicare not tested not paying for testing for unaffected people It comes under their statutory requirement that they not do screening and Yet for me for a family. It's not screening It's risk assessment to be able to tell someone you know what you're at 50% risk for inheriting a leaf romani mutation You don't have it the impact on that patient's health care And what it's going to cost to take care of that patient drops dramatically once you've made that demonstration So somehow I think we have to we have to create more than just Diagnostics and screening there's a third kind of testing and I think we need to get that across somehow It's it's easy. I think To get a payer to pay for the genetic testing in a cascade way. That's that that's a cheap That's What but we're we're not talking about leaf romani. That's not a Medicare age Thing is it? I mean you tell me. Yeah, you'd be surprised. I would yeah But but but I mean the the the issue is the patient who's sitting in front of you who has a disease And you want to know the the mutation to look for I'm allowed to use the word mutation in the next set of in the next set of first degree relatives And it's the first degree relatives getting that paid for is less of a burden and getting that initial test I can't I can't help but asking a cancer question because that there's a dimensionality for cancer in economic modeling at least in principle that could lead to different the very different perspective than the one That was just postulated about longevity and prevention and that kind of thing which is the the cost of of treatment and the and that you know zooming upward costs of Therapies so the and I'm sure most people know that the notion goes that I mean we already know that there are some genetic findings that will Specify the non-use of expensive treatment and so k-ras has been there for a while But now there's even the possibility. We don't know it's early days But there's a possibility for example that mutational load Could could emerge this way so a low mutational load might Portend a lower likelihood of responding to immunotherapy, which is an astonishingly expensive Therapy we don't against early days But if that kind of thing happened then you could start to imagine a scenario where actually the intervention of a genetic test could lead to a cost a Measurable cost savings to a proportion of patients now. These are hypothetical arguments To my mind it would be a very interesting To see a robust economic model of various use cases of this Are you aware of such a thing and this is something that is happening now? I realize that the field is evolving almost in real time and and but I think now one could there's recent data Sort of negative data on immunotherapy. They came out of one of the trials over the summer which now makes one think okay Well, maybe it's not going to be just blase and no one should die without getting immunotherapy drugs So maybe now one can start thinking about these kinds of questions in a way that where economists could could play an important role I think you go on top of that there was a presentation at ASCO this year from Neil Mason showing that On on average the at the average patient had or so the average per patient was about $8,000 of extra cost because of the toxicity Often this is out of pocket cost as well as costs are not covered for the health system in addition to the $126,000 for the drug and etc. So there's all these other elements there your risk benefit We talked about benefit in oncology. We kind of ignore risk now We need to get to the point of really having these things. So it's an exciting time Panel and then I think Josh's hand and then Tony yeah to directly answer the question I'm not aware of any studies, but there are some interesting models like the cistnet model in colorectal cancer where you could presumably You know for those individuals in that model that develop Colorectal cancer could begin to Test out what would happen because that takes into account all cause mortality and all the other things that would in fact impact You know longevity I mean that the the other side of the coin is is that most of the targeted therapies that we're developing now We're not cured of therapies And so what you're extending life, but at a very high cost that you for some of them Do you have to continue to do so I think it's it it would be really interesting to extend modeling into that space because it might help us to give some clarity around You know if related to toxicity issues related to longevity of treatment death from other causes, etc Could potentially provide some clarity. It would be a gnarly modeling problem And actually that's I mean that's one of the the the roles of pharmacogenetics It seems to me that to to identify subsets of patients in whom cheap effective therapies are the way to go and Other subsets of patients in whom you know that the cheap effective therapies are not going to be the way to go So use expensive ones. So even something like clopidogrel is Conceptually the same thing as the what you're talking about. I think With with with with four or five zeros less than what you're talking about With Josh Stantonia and then wrap up unless anyone has any other burning questions with Josh I was just going to mention that I think a number of people touched on this But one of the aspects of economic evaluation that ignite would be well set up to study is is the behavioral aspects of it so response to economic stimuli like out-of-pocket costs and And that can create sort of a feedback loop and we see that in the clopidogrel example with sort of the out-of-pocket costs related the alternative drugs where not all patients can afford to switch even though they should and and then You know, I think in general we just need to be More aware of behavioral aspects of the models so that you get all your information from clinical trials You're not gonna be modeling the real-world behavior of patients So we include things like statin discontinuation after an adverse event in our statin model because that's actually what we see in observational studies Add to that and that is that one thing we haven't talked about Probably to good effect is personal utility and the I saw it on one of the slides It was presented earlier today with the idea that this is important information for me And there's another economic tool that's emerging called value of information analysis that I think this is also a really Interesting tool that could be brought into some of the work that we're doing to just say how much is this information really worth to you Over and above what medical benefit we think might accrue from it Just I just wanted to just thinking back toward this discussion Just want to go back to Mark's comment in response to my comment about about dead people not incurring health costs and just Just just kind of bringing a reminder of of what came up several times during the parent meeting and earlier today Is that cost savings is not the only thing that economists are focused on and so yes You know certainly getting a diagnosis and getting its kids IEP that that incur is expense But hopefully in the you know and and and maybe we'll save money later But in the immediate future it's gonna it's gonna improve that child's education, which is gonna be a good investment So just keeping that in mind Thank You Tony Ned you have some Yeah, I was happy letting it go until Mark made his last comment. So I Think the issue about personal utilities fascinating and something that the field I Believe the field is gonna have to figure out at some point because I don't have any debate that there's personal utility to the information My debate comes into who should pay for it should be paid for through health insurance. I Think that is not a settled issue Far and away. I was fascinating listening to the comments about it's hard to Put value on or the insurers are having trouble with the value of reproductive information that informs reproductive choice For my standpoint in the personal utility you should be able to figure out a way to put a Value on that that would put it in the we should pay for it, but the other issues about The health value of other things like getting the diagnostic Odyssey just knowing getting a diagnosis I think that's a debate that It would be nice to have some evidence around how it translates to an important health outcome Yeah, I would just say I do not Interpret my words to say that personal utility should be paid for but I think we're on agreement that it is an interesting research Question and I think that ignite would be well positioned to begin to look at that Our briefly in the more closer wrap up so but but that to that point I mean what we do see at least in terms of the diagnostic Odyssey cases is that personal utility Even when there isn't a diagnosis they stop going from place to place So there is actually a clinical utility and a cost savings by having the personal utility fulfilled So I don't know how you how you balance that yeah, I mean there is there is analytic data looking at the Utilization impact of medical uncertainty And that is quantifiable and it is a question that I think an answer would be relevant to a payer But that analysis has not been done in the context of ending the genetic disease diagnostic Odyssey So that would be that's something else that's on my list of interesting studies to do So wrapping up a couple of things that really stuck in there One is about the high variability and arbitrary nature in in third-party coverage And I think you know there's there's things that can be done just in terms of of cataloging that Once we identify how much variability there is and you know Bob lives that but it's not really in the public domain How much variability there is and so that's one opportunity I think to really get in there and say here's the problem Now we can think about is it a problem with fixing? How do we fix it etc? another thing is around the The idea that models that are good enough, you know typically Most of us in the room have been drawn to this point in our career by doing Academic endeavors of some type whether we're at a company or a university or whatever it might be and and typically there's a level of Rigger that is necessary to get into New England Journal of Medicine that is different than being good enough Rarely is good enough unless you're based in Boston you good enough doesn't get you in New England Journal of Medicine So, you know, sorry little of Boston joke there Right so the idea that that we have this practicality around applied economics That is different from our our egghead academic hat is is is really something that Is we're it's a good thing. We're at that point where we need that but it is different than what we're not used to doing And so I think we need to move on that Link to that is this idea of do we do economic modeling or economic analysis? And in the end of the day having the models from the analysis will allow us to go into other scenarios But too often we're doing modeling based on synthetic data or semi synthetic data as opposed to actual real data and The recently one of our our folks started working with the finance group and we can get actual data on almost everything We need and it's hard. It takes a little bit of extra work It does take the language issue and thankfully he had an MBA and could expect at least part pigeon of finance But the the idea that one could get real data rather than averages from some from Kaiser Is is you know things that can be good guys is great for you But for everyone else it may or may not be representing what we need and so there's there's some opportunities there you know we talked a lot about the the Diversity issues and disparities and such and I think this really comes into to at least on the idea of of the the poor Often the folks that that not only can they not afford the test, but they can't afford the extra day off I I still remember that very first woman that I took care of that Told me that she had to quit her job in order to be receiving cancer care Because she couldn't they wouldn't allow her the days off to come and be treated and that was stunning to me And and still sticks with me that there are a large number of Americans Who getting a day off for care much less? Pain for it is a big deal And so we need to be thinking about that because of the large numbers I think understanding In what ensures want and and at least acknowledging these endpoints if not, you know preferably adopting those endpoints is really important And there's been some efforts by ignite to do that. I think more needs to be done Because it may be a one-year endpoint for something like increased adherence That that would be enough for them to move the needle as opposed to Something that needs to be your 10-year survival that we all I'll care about There's also some changes in the health plans that I think are are worth us looking at There are now a number of health plans I call them gatekeeper health plans coming back to really the primary care physician as a gatekeeper of the money as and we're seeing patients now who Had to to find another way had to switch providers in order to get treatment for Their their cancer because their gatekeeper didn't want them to and made statements like well, you got to die as something And and so the idea that we're we're we're having such a variety of health plans That a bit our aura is an important part and then last little point I'll make is is that All of our institutions those of us that are health care institutions we all have coding experts and as Testing has become a bigger deal and tumor testing at least at our place is what tipped it We now have folks that are spending energy to determine How do we get paid for this and they are finding ways to get paid for it legally? and and and so You know some of this is starting to shift as it becomes a big enough deal and so I think you know the part of it as the field becomes bigger than Maple syrup disease in a few patients and now it becomes the whole pancake We we got a lot of people that are now starting to care So I think we'll stop at that point. I want to thank the panel and the speakers I thank the audience for the questions and turn it over to ebony for the weather. I Would like to thank the speakers moderators and discussants that Talked to us throughout the day. I think it was really great presentations really great discussion and it led to a lot of Recommendations we are now up to the break what we're going to do is the planning committee will meet next door We're going to put together what we've heard throughout the day I am requesting that the moderators of each session also join us to make sure that we've captured what the recommendations were for your session and We will meet back here at three fifty five and Chris is going to lead us on the review of the recommendations to make Sure that you all can give input on us being able to capture what we've heard and Katrina and Lon will then Lead the session to prioritize those recommendations. Oh Colette has some housekeeping The is this one the smaller meeting room is number double g 607 it's down the end of the hallway on the first floor closest to the back doors Oh, so we're not meeting there. We're meeting Okay, so just follow me or follow Colette. You see I Don't