 Okay. Welcome to the session three of the meeting. The session is going to focus on the logistics of population screening. I'm Carol Bolt from the Jackson Laboratory and a member of the Genomic Medicine Working Group. And we're going to start off this session with Melinda Massart from University of Pittsburgh Medical Center. And Melinda, turning it over to you. It's really an honor to be here today and to talk about opening the floodgate of results and are we ready and how will we handle this in health care? As you can see, I am a family medicine physician, so that probably gives away the punchline of what I'm going to talk about today. But I do want to just sort of acknowledge that I think, you know, I may be posing more questions than answering them in my session today, because I think there's a lot we still really need to think about. So I think we all recognize that there is a tipping point. And that's why we're here today to really be thinking about the difference between population screening versus risk-based screening. And in our primary care precision medicine clinic, we have the great opportunity to be able to do very extensive primary care-based pedigrees with our patients. And although patients are coming to see us for an indication, we often find one to four additional genetic indications once we go through that pedigree. And even with this level of detail, we know we are missing folks. And so in our office over the last year, we have started offering population-based screening to our patients, even if they don't meet specific criteria for indication-driven testing. And I think that's really because the cost of testing has come down so much that we are even able to consider this. However, this is just on a very low, small scale. So how do we really think about when is the tipping point in the larger scale and going nationally with population screening? So a couple of things I want to talk about first before talking about how do we handle the results is, one, this concept of democratizing genomic testing and using genetic testing as a tool. And someone brought this up earlier, even, I think, using some of these exact same examples. But really, genetic testing is a tool now. And we really need to think about how to scale that and put that in the hands of all clinicians, just like radiology is a tool. Cardiology is a tool. You know, when I order an MRI or a CT skin or an X-ray, I don't refer my patients to go see a radiologist, right? I order those tests myself. And same with many cardiology tests. If I want an echocardiogram or a stress test to restratify someone, I order that myself. And then I take that test result and, based on the finding, refer to the appropriate specialist to help manage that particular situation. I think we need to start thinking about genetics and genomic testing in that same concept. And many of us have acknowledged the challenge in scaling up access to genetic counselors, which I think also is probably an impossibility when we start talking about population level screening. And so, again, how do we restratify patients and then get the right patients to the right level of care after that testing? The next concept I wanted to just point out, and, again, this has also come up, is really thinking about a single test future. You know, one of the challenges right now around democratizing genomics is that testing is so highly nuanced. And this is really a barrier to most clinicians being able to utilize genetic testing or screening. But if we move towards a single test model in the future, then we can also think about how does a single test be applied across the lifespan? And when is it appropriate and relevant to unmask certain results? Even if we have them all, initially, we don't have to interpret and analyze them all initially. And we can think about appropriate times across the lifespan when they are relevant and appropriate or when there's an initial clinical indication. And in the future, this should be able to happen really in seconds, right, when there's a clinical question at the point of care. So what do we need to achieve population scale genomic screening? And, again, lots of folks are talking about these different elements today, and this is certainly not exhaustive, but I think we have some critical ingredients that are needed in this recipe. We need national buy-in. We need community and foreign processes. We need integrated clinical decision support for management, informatics, infrastructure, and educated workforce, simplified testing, data sharing mechanisms, patient empowerment, enhanced protections through Gina for privacy and security, and funding. And if we are able to achieve this, who and how would we handle all of these results? So would testing be centralized like newborn screening within state labs? Would health departments be responsible for notifying positive results? What would actually trigger the interpretation at what stage of life? And what would we consider actionable and when? Would those results then go to the relevant specialty care providers? I think that this would be very challenging because this would require a chaotic network of referrals and often great delays in being seen as we already know is happening nationally in genetics clinics. And if it doesn't go to specialty care, does it go to primary care? So I obviously am going to advocate that, yes, the answer is it should go to primary care. And I know many of you are thinking about this as well. I think the primary care workforce makes the most sense. We truly are the orchestra conductors of health. Primary care includes pediatrics, family medicine, internal medicine, obstetrics and gynecology. We are the first line of medical care and have the lowest access barrier. We are available across geography. We provide care across the continuum of life, the age span, it's multi-generational with broad scope of practice. And we are the home of preventative medicine. We have multi-disciplinary care models that already exist, including pharmacists, nutritionists, therapists, social workers. And now in our clinic, we're adding genetic counselors to explore what this looks like. And patients have honestly already expressed and it's been documented that they have a preference for keeping their genetic concerns within the primary care space. Primary care also, the scope of practice aligns with genomic screening, right? So in primary care, we do preventative care, which are risk tests and panels. We do prescribing management, which is pharmacogenomics. We do routine cancer screening, which is genetic cancer risk assessment and multi-cancer early detection technologies. We do prenatal care, which is prenatal carrier risk and NIPT. We do newborn care, which is following up on those newborn screening results. And we do chronic disease management, which will be in the future polygenic risk scores. Also screening already lives in primary care, right? This is all the national guidelines around screening. And this is already done all in the primary care space. So is the primary care workforce ready for this? No, they are not. I mean, I want to be positive. It's not an F. I did not give them an F. You know, there's a lot needed to make this happen. And, you know, I think the next question, though, is primary care able to do this? And the question is emphatically yes. I absolutely believe that primary care can do this and should do this. So what do we need to do to prepare the primary care workforce? Some of the critical needs that we have are time-saving efficiency measures, right? Time, time, time. There is never enough time in primary care. Knowledge, we already know that's a major barrier. Confidence to manage all of this, both on the side of the providers and the patients. And really, I'm going to lean heavily on robust informatic infrastructure to support data integration, reanalysis, curated updates and portability of results. So possible solutions, we need clear and concise just in time clinical decision support to manage results across the lifespan. Clinicians in primary care are never going to have all the knowledge needed. They just won't. They cannot add it to their already very full plates. So they have to be able to lean on clinical decision support and trust that it's up to date and have the confidence that it's going to support them in the algorithms needed to manage all of these screen conditions so that we're actually doing something with those results. We need minimal viable product for supportive management and counseling. Primary care providers should not become genetic counselors. That informed consent and guidance is really the secret sauce of genetic counselors. We should not ask primary providers to do that and they don't have the time to do that. So what is that minimum viable product that they need to do informed consent to be able to integrate results into the healthcare records and into the care and management of their patients and then when to refer those patients up to that next level of care. And of course we need enhanced referral systems for management beyond primary care and better electronic health records ready for the needs of genomic medicine. I think this has already been acknowledged today, but our current electronic health records don't even help us manage diabetes at this point or routine cancer screening guidelines at this point. How are we going to ask it to add all of these additional components? I don't think that's a defeatist kind of thought. I really just think it's a challenge that we have to elevate these EMRs to do what they truly can do. And finally, you know, community informed models are necessary to prepare the public to get to this level of advancing genomic screening. We need to think about who are interested parties and there are many interested parties that belong on this list, but the two I really want to highlight are the community or public themselves and the clinicians who will be out there doing this work. So how do the key interested parties want this to happen? What should the models look like? Are they federal, state, local, regional? What models will be acceptable to the public and readily adopted? You know, are they going to be universal or population specific? And how do we ensure diverse and equitable uptake of the models across the population within the U.S.? You know, we talked about these classic screening criteria earlier today. And I guess the question for me every time I read these is who actually decides these answers, right? They kind of pose the questions, but who decides the answers? And I really think that needs to be discussed and engaged with the community, both the population at large as well as the primary care clinicians who will implement this. And then finally, I'm going to strongly advocate that we not do a fire hydrant model, that we don't prep everything and then just release this massive onslaught of results and information, but instead really think about proposing a trickling faucet model. We pick one high value, high evidence screen. We AB test this out in different mechanisms in different places with the community, both the patients and the clinicians. And then we layer on additional screening tests when pilot phase is deemed successful. And with that, I'm going to pass on to the next person. Thank you. Thank you, Melinda. So next we have Peter Kraft. Peter makes his way. I'm going to talk whom to screen, when, and how. Thanks. It's really great to be able to join this meeting. I've caught the last five minutes of the last session. It was already an exciting discussion. And first a disclaimer, I'm a statistician and a genetic epidemiologist. So I'm primarily interested in gene discovery and gene characterization, so estimating penetrance largely in the general population. So I think about clinical testing and genetic testing both in the clinic and public health, but I, most of the clever things that I have to say about it, I've learned from reading articles, we're talking to experts like many of the folks in this room. So if I say, manage to say something clever, pat yourselves on the back. If I say something foolish, that's entirely my responsibility. So I wanted to start with just reviewing a successful non-genetic screening program, so the widely accepted mammography screening for breast cancer. So, you know, the thing to note about this is it does vary across time and across different contexts, different countries, different public health systems, but there's a general agreement that for the general population for an average risk woman, they should start screening between their late 40s or early 50s. And this is based on a balance of sort of risks and benefits, both to the individual and to the health system and society as a whole. So, I mean, it fits the classic screening criteria, so it's an important health problem. There is an accepted intervention treatment. We know something about the natural history of the disease and I think importantly for this context, the case finding is definitely not a once-and-for-all project. The guidelines call for getting repeated mammography. It's not just you show up. We don't find any evidence of cancer in your breast today. Congratulations and have a nice life. We're screened repeatedly every two or three years. But because a lot of the guidelines are aimed at the general population, there's still a recognition that there will be some people who are at higher risk than average risk who we might be missing. And the question is how do we identify those folks and clearly genetics screening, genetic testing is one way of doing that. And the US Preventive Services Task Force has already made this an area of important research, something they're looking into. So I want to talk a little bit about genetic screening and how it might help in this context. But before I get to that I just want to flag the sort of coverage or the uptake of mammography screening. So it's not enough just to have a set of guidelines that everybody thinks are good guidelines. You have to implement them. And even for something that is pretty uniform at age 50, start screening every three years, there are still coverage gaps. And these are influenced by a number of factors. It could be social and economic status. It could be distance to the nearest screening center. In the middle panel there is highlighting immigration status or time since moving to the United States as a potential barrier. So all of these things are should be kept in mind. We've already heard about a couple of those in the previous talk and I'll come back to this again. So when thinking about the potential utility in a population level genetic screening, we should start with the baseline which is sort of the current guidelines or the current practice. So this is a schematic that's sort of describing genetic testing, clinical genetic testing for three tier one CDC conditions, hereditary breast and ovarian cancer, Lynch syndrome, and familial hypercholesterolemia. So and basically the story is that we catch folks and flag them for testing based on a personal or family history of disease. So those bottom three arms there, no pointer. So the bottom three arms are for folks who we've identified should be tested. And of course if you're tested you might end up having a pathogenic variant. You may have informative negative test. So we sort of know what the genetic variant that's segregating in your family is and you did not test positive for that. So we're fairly reassured that you don't have the high risk variant. But a lot of folks end up in this middle category where either you test positive for a moderate risk variant. So I think check two in the case of breast cancer. Or it's uninformative. Like there's something going on in your family but we haven't been able to pin down what the genetic variant is. And this is the setting where I think there's been some, there's actually some clinical invitation now. Some companies are offering some clinics are offering polygenic risk scores to help differentiate those moderate risk folks. So for example taking that check two example. If you didn't know anything about the polygenic risk score or other risk factors you're sort of right on the bubble of guidelines in terms of MRI screening. But if you had additional information about the polygenic risk score or other risk factors you may end up being fairly comfortable that you're actually well below that or well above that threshold and you could take action accordingly or be. So that schematic that I was showing oh and the other key thing on this schematic is of course the folks who we didn't test because they don't have a positive family history or haven't been diagnosed with the disease yet or for other reasons. We haven't been able to get them to the testing clinic. So there are folks who are walking around with a variant who haven't been identified. So this missed opportunity. So you know that little schematic was already somewhat complicated and when you go to the actual guidelines they're even more complicated as I don't need to tell many of you. So you know before we even get to population screening there may be an intermediate step which is to automate some of the tests to really identify people like actively go out and look for folks using EHR or other records to flag people who should be screened. And my colleague at the NCI, Kirtina Goddard, has implemented a pilot study charm which is the cancer health assessment and I can't read my handwriting so RM. Which basically takes sort of the typical process where people are sort of opportunistically identified and makes it a little more systematic and they were able to show that this was able to get more people in for screening who should be or for testing who should be according to guidelines especially among folks who are underrepresented. So moving to the other scenario where we did undertake population screening for these three conditions. So now everybody gets tested and you can end up in sort of three bins. The middle bin there is folks who would have been discovered using sort of current care. And you can compare these two arms. What would happen if we kept things the way they are versus what if we implemented genomic screening and in this one particular case, there's one particular paper where they did a simulation model, it turned out that the population screening was able for these three diseases together, was part of the key argument they were making, was effective. You were able to identify and prevent more cases of cancer or deaths in cardiovascular disease. You had a better quality adjusted life years and what's particularly interesting or relevant for this session or the title of my talk anyway is that when you do the screening or when you, maybe you did the testing early but you've unmasked these results at different ages, 30, 40, 50, affected the utility of the screening program. So in this case, starting screening at 30 years old was the most effective. They did look at what if you started earlier, 20 years old or so in this particular setting. The gains were marginal and we're not necessarily offset by other factors. So starting at 30 made sense. So I'm just going to come back to this picture and mention some of the coverage gaps that are, I should have changed the title of the slide. The title is aside but the point is I'm thinking about the gaps that might exist for population genetic screening. So there's going to be costs. There's going to be how do we cover barriers in terms of transportation, getting to the testing center, the time to undergo the testing and there's going to be a burden on the healthcare system in terms of returning the results and the subsequent follow-up. So all those things will have to be considered. How am I doing on time? Good. And then I just, you know, when I was reading the Wilson and Younger article retrospective of the article in preparation for this meeting, it really made me think of the Jeffrey Rose 1985 in our article, Sick Individuals in Sick Populations, where he talks about sort of two strategies to lowering the burden of disease. There's the high-risk strategy and the population strategy. So the high-risk strategy, I think sort of what we're talking about, let's identify that people are at higher risk and let's intervene on them in the case of cancer screening early, if we catch the disease early, versus the population, which is about shifting the underlying risk. If there's an environmental exposure that's driving a lot of the population burden, let's change that. So the high-risk, the individual gets a big benefit from that intervention, but most people will not benefit because we're focused on a small proportion of the population for any particular disease, I should note. Whereas on the other hand, the population, you have actually a big population impact, but the difference for any one individual in the population might be small. So this is another slide I borrowed from Katrina, sort of making that point contrasting individual level interventions on tobacco control in this case versus sort of population level, and you're sort of getting, there's more of an impact for the broader population approaches. But I guess the one thing that I wanted, the point I want to make here is we shouldn't forget about those other approaches, especially when we're thinking about complex diseases. So again, breast cancer, there's lots of, it's multifactorial, people get breast cancer for all kinds of different reasons. It's a small proportion of people who get breast cancer because they carry pathogenic variants and they don't get breast cancer. So there's a potential for intervening in other ways. But we should be clear that it's not an either or, right? It's a both and. Both of these strategies can be in play at the same time. So thank you very much. Very good. I believe our next speaker, April Adams from Baylor is online. All right. Can everyone hear me? Yes. Okay. All right. Well, thank you for having me and accommodating my need to be virtual. I'm April Adams. I am an assistant professor at Baylor College of Medicine and I'm primary clinical and I work as our reproductive geneticist. So today I'm going to talk to a little bit about addressing the challenges of genomics screening in populations underrepresented genomic databases. So I'm sure that I don't have to prove to the audience here that the U.S. is a diverse population. This is just an snapshot of the race and ethnicity prevalence by state from 2020. And you can see it kind of goes from the most highly represented group over to the second third and then kind of a diffusion score of the more of the populations, the lower prevalence. And interestingly, so when you look at this data and you compare it to 2010, you can see that our reporting at least or how people identify and the admixture of people that live in the United States has definitely changed since 2010. So the largest group represented being white identifying is not Hispanic comprised 57.8% in 2020, but that was down from 63.7% in 2010. And then you can look at this in another way looking at over the different age demographics in the United States and you can see that also that diversity is reflected in younger populations as well. But the thing to consider with self-reported race and ethnicity is that it can often be pretty unreliable and it may not capture the true diversity of a population. So when you're thinking about what is our actual genetic diversity look like, what you find is when you look at people across different populations in different locations with different cultural backgrounds, you see that those processes impact genetic diversity significantly. So even a population next door to each other with a different cultural practice may have some different rate of changes in their genetic diversity and this can't be reflected in our race groups that are categorized pretty broadly as white, non-Hispanic, black, non-Hispanic, Hispanic and then we categorize everybody else into this other group. So just to kind of set that stage up, we are diverse and we're probably not even really capturing how diverse we are. So our identity is multi-dimensional. So our identity is our ancestry, our genetic lineage, our genealogy, all of those things that we have no control over and then it's still also impacted by population movement and cultural practices and all of that. But then we also have these other categories that are placed on an individual such as race, which is the construct that we create based on physical attributes or things like that that can also change over time and depending on where you are and where you live. And then ethnicity as well is an area in which it's a construct that really looks at your language and religion and nationality and it can be self-reported but it is also something that can change and also people may not know enough about their prior history to really determine their ethnicity and there's really no consensus on what race and ethnicity should look like and so I say this all to say that we have to be very thoughtful about how we look at race and ethnicity when we are looking at clinical implementation and designing research studies because it's information that we may be building upon the bias that already exists there. And I say that looking at okay how are we doing in representing people from diverse backgrounds? So we're probably under-representing our diversity and then we're also not even capturing the diversity that exists. And so when we look at some large studies and large databases we see that we've made some improvement but overall it's still heavily weighted as has been previously mentioned towards individuals who have European ancestry and that also is going to still limit the amount of diversity that we're seeing in those studies. And so this lack of representation in addition to our inability to really give people the appropriate categories when we're looking at ancestry, race, ethnicity leads to this perpetuation of these health disparities that already exist which is really the opposite of what we want to do with genomic screening and sequencing. So what are the drivers of these disparities? And a lot of the drivers of these disparities have to do with the social and cultural context in which we live and so looking at things like social determinants of health and how do these experiences lead to our health and well-being or to decrease life expectancy higher healthcare costs and things of that nature. What we are missing is we are lacking on that diversity of people included in studies and when we do that what we find is that you're going to also miss out on the genetic factors that are interacting with our environmental behavioral and social determinants of health to lead to that disease. And so we know that the parts of genetic diversity that are going to flow through with our population and cultural processes without including those people we are going to miss out on figuring out where we can make that in actual impact in health. And I'm sure that I don't also have to explain to the audience that there are clear disparities in our health outcomes and just a couple of examples of when we look at race and ethnicity obviously with the caveat of we're mixing race and ethnicity together here which may not be the same thing we see that there are big disparities in groups specifically in non-Hispanic black individuals the rates of death, secondarily to diabetes and heart disease far exceeds the individuals that identify as non-Hispanic white and so how do we make that gap smaller if we are including people in studies and not only does this disparity impact adults this disparity actually starts in utero, right? So if you look at fetal mortality rates, the rate of birth in non-Hispanic black fetuses is twice that of our non-Hispanic white population and then this still translates to the same when we look at infant mortality as well. So with that I just wanted to take a case example of looking at reproductive screening and the reason why is because it's an interesting area of screening. It's an area in which a lot of people are captured but it's not really done in a population health sort of manner but when people are adults pregnancy may be the first time they actually encounter a genetic test and so with carrier screening it's really looking at can you decrease the morbidity and mortality of fetuses and infants by screening asymptomatic individuals for autoresomal recessive and X-link conditions and ideally you want to do this in the preconception period and give people the opportunity to make reproductive decisions based on that information. And traditionally this was done based on ancestry or in a pan-ethnic manner or based on family history but it was limited because it only captures a small amount of diseases that may impact a fetus or infant and it also as we mentioned people don't know their history or their ancestry or very well and so you're excluding a lot of people who may actually be at risk. So looking at that and looking at as we talked about there's different ad mixtures, patients that can't identify their ethnicity you look at something for example like sickle cell disease people really targeted African-American patients which makes sense but then you're missing lots of newborns who don't identify that way who also will end up having this condition. And so this is what led to a more expanded carrier screening approach. And when we look at expanded carrier screening we're going to be able to look at more diseases do a pan-ethnic survey and really try to capture an entire population and then what we found is the more diseases you screen for the more people you'll find our carriers which is great. And so then we get these guidelines that say okay these are the conditions we should be screening for right so frequency carrier frequency of greater than or equal to one in 200. So, hopefully let's make it equitable and all pregnant patients of those planning of pregnancy should be offered at least this tier 3 or a carrier screening frequency of greater than or equal to one in 200. However that is a little bit more difficult in actual practice. So when we look at some of the challenges in expanded carrier screening I really want to focus on the fact that one we are counseling people who may not be represented in a lot of these screening panels and then we really don't have the right access to do this at a population level. So just an example of a study that a systematic review that looked at carrier screening research studies and in this study what they really found was that it's been curious we needed a great job however these were small studies and even in these small studies they only had a very small percentage of patients who identified as non-European ancestry. So there again you can see people are just not being represented and not recruited into these studies. Additionally when we talk about access we have these huge barriers in access when you see that patients who identify as non-white are going to have a harder time accessing care due to costs related to care and they also have higher rates of being uninsured and so that makes things that are like carrier screening which can run anywhere from $200 to $2,000 very cost prohibitive. And then another piece that is also going to be a barrier to care barrier to access for patients is going to be provider bias and discrimination and what you see is that not only are patients having barriers to insurance and all of those things they are also seeing barriers in getting the same care offered to them that would be offered to somebody who is a part of the majority. So they may be less likely to get a referral to the genetics clinic they may be less likely to get a genetic evaluation or to even be offered a screening test and additionally they may not access care because of prior experiences perceived or discrimination and poor treatment in those facilities. So when looking at our criteria for the population based screening I think some things definitely are already there but some especially are can we offer this to anyone who wants it no a lot of people still have a lot of lack of healthcare access so there is a big lift to figuring out how you can offer this in an equitable way and can you are we actually offering it to individuals in a routine in an equitable manner because many times patients may not be offered a test just based on their providers perceived bias about their decision making their race, ethnicity all of those factors and then obviously as we've discussed lack of representation in genomic databases really does lead to difficulty in counseling and less downstream research into how do you mitigate the outcomes of positive results but all this is not theoretical there are many many carrier screening tests out there you can also do preconception genome sequencing exome sequencing and so the cat's definitely out of the bag and people are being sequenced we just are a little behind in figuring out how do we implement this in an equitable way and so right now we're in this phase of kind of having this increased gap in who can get it who can't and who's benefiting from this process so in looking at that we took kind of a step back and said what are the things we need to be thinking about when you're going to provide equitable genetic services and kind of looking at the NIMHD framework for equity and looking at the different domains and how do you impact the lack of knowledge about genetic variation and all those different levels of engaging individuals, their families their communities, educating healthcare providers building trust in healthcare systems as well as making sure that people have access to culturally sensitive care to affordable care and that their you know wants and desires are incorporated into how we disseminate this kind of care and so just moving on from that looking at what are some principles of equity when we have patients who may have been who may be underrepresented who may have a marginalized population really focusing as we develop clinical implementation strategies and research protocols incorporating person-centered models to help empower marginalized individuals and communities and I know we're going to talk more about that in later sessions acknowledging historical harms and using that knowledge about those things to build better systems right so when we talk about things like AI and telehealth and all of that making sure that we are not building that same bias that structural racism into these systems and having that thought before instead of an afterthought of oh now we realize that this is a disparity let's try to fix it later really having respect for individuals choices and not creating shame around their choices and their decision making and being creative about meeting people outside of the healthcare system because sometimes that first step is getting people to trust the system and actually enter it and then a big piece is also looking at how do you better support health literacy language cultural context because uncertainty for one person based on their language cultural background experiences may be very different than the next individual and understanding that there are definitely first vote burdens and being able to do all of that so the gaps that I think need to really be addressed are one increasing diversity and inclusion in the workforce because a lot of these principles of equity can be done when you really engage what patients need and patients tend to do better and feel less marginalized when they see that there's not only representation in the patient population or the research population but in the stakeholders of people sitting at the table and making decisions and making policies also identifying and limiting barriers to participation as I mentioned so maybe just getting out into the community being engaged in understanding that there are going to be multi-molder ways that patients need to be engaged and care and not immediately thinking oh they don't want to do it because they didn't show up because maybe they needed a ride there and incorporating the principles of equity in all levels implementation it really should start with the hypothesis right it should be from the very beginning the first question that you have how do we create this in an equitable manner and then the big thing which is a really heavy lift is how can you expand beyond race how can you incorporate social determinants of health with ancestry into these research questions and so that people can be moved outside of those boxes to really understand what are the drivers of their health outcomes and thank you thank you and so our final speaker in this session before the discussion is Kelly East from Hudson Alpha all right good afternoon I'm Kelly East I'm the vice president for education at Hudson Alpha as well as a genetic counselor so I'll be kind of wearing both of those hats in this talk and I really appreciate the opportunity to be here and be a part of the dialogue and share some of our stories and data from our experiences engaging both providers and patients in population screening initiatives and so one of the things kind of as I think about population screening there's some wins in the education space in terms of barriers that by deploying things on a population level there's there's a few less things that a provider necessarily has to know about their decisions around to get patients access to genomic care but at the same time population screening requires more education and knowledge and skills and confidence for the patients who are getting that testing to have the maximum amount of benefit and the least amount of harms involved and so I'm not here to I mean everybody in this room I think would agree that more education and training is needed but what I'm hoping to do is to share some of our stories and some of the themes that have come out and emerged as places where we can do more and themes that should be a part of the interventions that we need to deploy to provide better education and training around genomic screening for these audiences and so just as a bit of a context setting some of the studies that we've been engaged with that I'll be sharing some data from there is a cancer risk population test that we've been doing for a number of years so Hudson Alpha is in the northern part of Alabama so from a context standpoint the patients and providers we're engaging with it is the southeast part of the United States where that issue of access to genetics care is kind of at an increasing you know pinch point but this is a population consumer directed test that we've been implementing where consumers can go and self-select to have the testing and then we go and engage their providers but it's a cancer risk gene panel that is definitely meant as a screen not as a diagnostic test. Another study that we've been doing is the Alabama Genomic Health Initiative and I think we'll hear some more about this tomorrow from Dr. Corf but this is an array based test that we've been doing again for a number of years for individuals in the Alabama for adults initially it was actionable disease risk and then more recently we've added some pharmacogenomics to this study as well of note from 2017 until COVID in 2020 we were operating under a at what we call our population cohort where we were embedding recruitment sites out in the community trying to engage Alabamians in this in the study but then after COVID we relaunched as a clinical cohort where we started recruiting patients out of specific primary care offices so it was much more integrated in the patients clinical care and then finally there's a this is another study that we've been engaged with called SouthSeq which is not a population screening test but the reason I included it here is we've got some really interesting data that I think is useful as we think about educational needs and misconceptions in something like population screening SouthSeq was a whole genome sequencing study for affected infants and NICUs although it was a very broad set of inclusion criteria it was we focus more on which baby shouldn't be enrolled rather than which baby should so we had a really diverse set of programs that were engaged in that and also specifically with this study we were interested not only in getting diagnoses for the programs but to test a result delivery model through non-genetic providers and doing a clinical trial around that for all of these studies as we thought about scaling up education and scaling up genetic counseling we we focused on how can we best use our precious resources of clinical genetics professionals largely in all of these studies the front end the informed consent conversations and decision making was not done through a individual interaction with the genetic counselor it was done in different ways for different studies so we don't have time to go into a whole lot of detail but we focused a lot of our education interventions around the return of results and thinking about on the back end once you've got results how do we manage them and make sure that the patients are getting the right downstream care so the way I want to frame this is just talk through some of our lessons learned and things that have come up that have framed how we educate our providers and our patients that I think we need to care take even more maybe this is one that we've been talking about already quite a bit today and that population screening is going to identify a lot of people with an unmet need for diagnostic testing I think we all know that there's so many people out there that should be getting diagnostic testing and they aren't for a whole variety of implementation reasons and so in a couple of our studies in particular and as I pulled this I checked myself over and over that these numbers were indeed the same in these two studies and they were but for the Alabama Genomic Health Initiative and that information is power cancer screening test both of those we included family history as part of the intake process so when patients were enrolling into the study or signing up for the test they provided some information about their family history and one of the roles of our genetic counselors was to go through and review this although we've gotten a little better with automating that as well but going through and using this as an as an opportunity to identify those patients that kind of regardless of the outcome of their test result there's some additional actions that should be taken based on their family history and it was almost half of our participants had some kind of what we considered a flag generally speaking those flags were things that came up that would have made them a good candidate for genetics evaluation or additional genetic testing another thing that came that continues to come up as we've been looking at results that have come out of these studies is that oftentimes the people who get positive genetic testing results it does not corroborate with the family history and personal history that they have told us about this is a chart from a paper that we published not terribly long ago where we went through and looked at all of our positive test results in our actionable disease risk in that population cohort of the Alabama Genomic Health Initiative and looked back and said well how many of them would they have been flagged was there something there that indicated that they were at a heightened risk of getting a positive result back and the answer was not many and it also the answer was it varies wildly by the gene that you're talking about there's some of the genes that we reported that corroborated family history or it was zero percent there were others that were a hundred and kind of everywhere in between but what this leads to are really surprising results that we can't necessarily predict or counsel or educate well around at the beginning before you have these results and working with our providers we did a fair amount of education and talking with our providers at the front end and then as they started getting these results a lot of them got really uncomfortable about thinking about having to talk to a patient about a genetic test result that seemingly is completely out of the the blue and it also brings up that issue that we've been talking about some today about what do you do with management in a unselected population and someone who doesn't have that personal or family history you know what is their actual risk for disease should they be following the management guidelines that have been published that have come from those high penetrant or more high penetrant families another theme that we've seen come out kind of over and over from providers and patients alike is this over interpretation of a negative result and the risk of that leading to false reassurance and particularly those people that have that personal or family history that getting a negative test result doesn't necessarily decrease their risk especially just depending on what the testing was and how good it is at picking up the known risk factors and so this is where I wanted to talk a little bit about that South Seek study where we've got some of the we've got some really robust data around provider errors, provider misconceptions which are really big opportunities for identifying places where we can do better in terms of education and the tools that our providers need and so in South Seek we took all of our the pro bands, the infants that were having whole genome sequencing and we randomized them to either get their results from a genetic counselor at their site or from a non-genetics provider most of whom were NICU physicians that we had done some level of training with and what that looked like was a half day live training also our genetic counselors had a role where we were writing the result reports for this study and going through and kind of creating a bit of a roadmap in those result reports that providers could use that has had a lot of the talking points in it that went beyond just this is what the result means that included some more things about what this means for the family and contextualizing that result for them a bit more than I think is typical but so the patients would get their results from one of these two entities and we audio recorded all of it which has created a really robust data set that we're just now kind of scratching the surface of and dreaming of what else we can do with this data but the reason that we were recording it was we wanted to be able to look for errors it started as a safety requirement for the study and then it turned into a really interesting research question and this is a kind of busy slide but the take home point is that non-genetic providers were significantly more likely to make what we consider to be a major error when we were listening to them we categorize them as major or minor depending on whether we thought it would have a profound impact on decision making so there was a higher percentage of major errors in our non-genetic providers but 92% of those non-genetic provider disclosures did not have any major errors in them of note we went through and did some thematic analysis of the errors that we identified and the one that came up the most kind of over and over and over was this over interpretation of negative results and there's some quotes up here from our non-genetic providers about them wanting to kind of think genetic testing is better than it is many of us in the genetic space know the nuances and know how much genetic testing can't find and these were in affected infants that had a suspicion of a genetic disorder and we had providers that would tell them that this ruled out genetics or something along that theme or that the future children are not at risk and so that's something that certainly was part of our education it was on that report that we sent back but this is still what was happening in those conversations and you've got providers and patients that are really over interpreting these negative results and I don't think this is unique to diagnostic testing and it's something that's going to be a major part or needs to be a major part of education for providers and for patients when you're giving back negative or non-informative results. We also have awesome data here from our inherited cancer risk testing where this was a survey it's probably hard to read but this was a survey that we sent out to patients after they had gotten their results and they had them for a while and this was a knowledge question of asking them how what does it mean if someone has a genetic risk factor for cancer versus what does it mean for a person to not have a genetic risk factor identified and the reason I point this out is that there's a conceptualization of that is much more accurate for positive results than negative so almost 100% of those patients are saying that they appropriately picked the right answer for a positive result but only 72% of people selected the right answer for a negative result and a notable 27% of people who filled out the survey who most of them did indeed have a negative result said that a negative result decreases your risk compared to the general population and that's problematic that even for people that don't have a family history getting a negative result is not going to take you down below the general population risk but the fact of the matter is that the correct conception for a negative genetic test result really is not the same for everybody and it largely depends on that personal and family history and it needs to be contextualized which makes broad education messaging challenging because you're going to potentially over an alarm or under alarm people with that messaging we talked earlier a couple of people mentioned cascade testing and I think that's something that from an education perspective genetic testing is a little bit unique when we're talking to providers making sure that we're calling out the fact that getting these test results the impact of that goes beyond the patient in front of them which is kind of out of scope and it's a little it providers are not necessarily thinking about or equipped to handle that kind of downstream testing or talking about those risks when we look at what patients are doing after they get genetic testing results back from a population screen and I think this corroborates with other studies as well patients talk to their families a lot more than they talk to their doctors about it and we have an opportunity I think with working with patients and providing them the tools and resources they need to really facilitate those conversations with family members in a way that that I think we can do more to provide those tools rather than expecting providers to do that and finally I think integration with clinical care is really important to both increase the access to testing as well as the potential maximizing the benefits of follow-up so this is looking at those Alabama Genomic Health Initiative enrollment data and the blue of these donut charts are Caucasian or European individuals and this is the difference in you both of these have thousands of people in these data sets in the population cohort of the study we had and we had people from all 67 counties in Alabama in the study but we were out in the community recruiting patients into this array based population screening test and despite our best efforts we still ended up with a fairly non-diverse data set meanwhile in the clinical cohort these were patients that are recruited through primary care offices so it's embedded in care and we've been able to strategically pick and partner with diverse populations in the state but we've had a much bigger uptake of diversity and a much more diverse data set after integrating this testing into clinical care but I'd be remiss not to mention that there's probably a whole bunch of people that don't have primary care doctors and so if you only go that route that you're potentially also you're creating some barriers to access as well but so getting the test getting people in the door but what you do with those results having that integrated in clinical care is also it makes it much more likely for people to get that benefit either the follow on testing further evaluation and then the care based on their results but this is where for those things to happen providers have to have the knowledge to have the tools to be able to do that this has got to go beyond providing CMEs for providers and thinking about the things that have already been mentioned today with EMR integration and AI and telegenetic counseling resources and all of that and thinking about from an infrastructure standpoint of how to make this as easy as possible and as leak proof as possible for patients to get the information and the care that they need so as I think about some opportunities based on the things that we've learned and things that we need to be paying attention to as we go forward certainly interpreting and communicating negative results needs to be emphasized the vast majority of patients that do genetic testing are going to get negative results and we need to make sure that we are care taking those and making sure that we're not treating them all the same and finding ways to contextualize them for the patients that are getting those results back thinking about program infrastructure that can help support that family communication and cascade testing that we're thinking about that downstream benefit and there's a lot of education needs in that not only for people to be aware of their risk but where to go to manage that risk especially when family members may live in wildly different places infrastructure as part of our screening programs to use that as an opportunity to catch people that should be on a diagnostic genetic pathway and figure out how to make those referrals easy and more likely to be done whether that's a referral or whether that's providers that are able to then add on those additional levels of testing and then finally scalable processes for clinical genetic professionals to provide support I'm very much a proponent of non-genetic provider engagement in population genetic testing I'm really interested in thinking about how we as genetic providers can become a safety net for those providers and become a resource for them in helping on an individual patient level providing contextualization and guidance but not being the one that is necessarily trying to use our knowledge and our resources as judiciously as possible but thinking about who are the patients that would most benefit from seeing a genetic counselor and how do we focus those resources together Kelly will have to call it there we're a little bit over time okay, sure, I'll leave this slide up for a second but these are when I think about the places that I want to be doing more research and I think we should be coming together to do more research if you're right here thanks so much so we're open for discussion and questions so I'd be interested in sort of Kelly following up on your last slide I'd be interested in hearing from the other presenters about sort of research agendas that can address some of the, we've had a lot of discussion of gaps right and barriers so I'd be interested in hearing some of your thoughts on research agendas that could address those barriers and gaps and maybe we can go and order Melinda maybe you could start us off yeah, so this is something I think about a lot and I think a lot has been published on the barriers and gaps and really would like to start moving the conversation towards solutions so we actually just held a day long session using Human Center Design to prioritize solutions and possible interventions for integrating genomics and primary care and I think that someone mentioned earlier putting out a white paper about these potential solutions and really starting to have teams that are collaborating across the country to test these various different solutions to really find out what works best in different types of care settings there's no one care setting in our country that is one of the biggest challenges that we face and so we're probably going to need several different models to find solutions that work successfully for various different care models but that's one place I would really like to see a lot of research moving forward great, thank you. Peter, any comments on it? Sure I mean two things that I think came up in the previous session were around understanding penetrance in the general population and I think April mentioned sort of being able to interpret particular variants that may be very rare in the existing databases but we don't those data is artistically representative so we may flag something inappropriately as being pathogenic because it's rare but in database but it's actually not that rare in some of these samples so I think those are still important areas to research. I do think we're filling in some of those gaps so certainly for the more common conditions that we might be screening for I think we are getting robust estimates of population penetrances and those were as biobanks and cohorts get bigger we're going to have better estimates of those and I mean Heidi can maybe speak more to the reference databases so I know Nomad just had a big update and there may be some interest in sort of better assigning folks using genetic ancestry as opposed to population levels or self reported race and ethnicity that might help with the interpretation and that goes to my other thoughts on research areas would be sort of on the informatics and both sort of how do we get the informatics in place so people can interpret these things and then when we are going to be probably maybe Kelly can speak more to this is we will be relying more on sort of automated reports as a way of sort of getting that conversation going and then maybe the clinician or the specialist would be sort of more answering questions that folks might have but there might still be some informatics and I just want to make a shout out to the the leaky the leaky tap so that part of the research will be the implementation as well like how do we do this most effectively April would you like to make a comment on research agendas that you think would be essential to addressing some of the logistical issues you identified yeah I think that you know everybody has made really good points so I think when we talk about how do we you know engage more people and utilize the workforce appropriately all of that I think really being thinking more out of the box about how we're getting out to other communities because if you look right now where all the geneticists and genetic counselors are they're in major urban areas so you know I think definitely focusing on how do we utilize technology better to be that consult or that safety net person for somebody's primary care home obviously that implies that people have a primary care home which is a bigger issue but utilizing those structures that work that's an example to grow that type of infrastructure or more and then I think that also just to reiterate I think that looking better at how we more thoughtfully design studies to look at diversity and how we are categorizing people because I do think that we are you know the genetics community is definitely aligned in a space to change that perception and help people understand better why ancestry matters and how we utilize it in a you know equitable and thoughtful way thank you Kelly any comments from you sure I I agree with everything that everybody just said and I think when I think the things that I'm most interested in and think would be really impactful is we're talking about thinking about how to pilot and test different models for interpreting and managing genetic test results thinking about how the combination of the testing labs the primary care provider and maybe a clinical genetic specialist that how those could come together to create pipelines that they get the providers what they want most which is an answer for what does this mean for this patient not what does this result mean but what does it mean for this patient and what are my next steps and thinking about whether that the combination of tools but also where the human brain can supplement those tools in a scalable way so one of the things I'll go ahead Terry and then Mark sure so Pete I was delighted to see somebody quote Jeffrey Rose it's wonderful to see his name again and I was just curious you gave some good examples of population wide efforts in smoking tobacco control what would you think would be population wide efforts in genomic screening or identification of genomic risk yeah so I guess I think of the population screening for genetic risk is falling more in Jeffrey Rose's high risk approach so it's identifying those folks who are at high risk so which we're sort of doing now we're missing as data there's folks who aren't being recommended to testing who would benefit from it so yeah I don't know if that exactly what this meeting is talking about is what I would think of as genomic screening Epidemiologists sticking together the rest of us are going who the hell is Jeffrey Rose so the question I would have for each of the speakers is I think you've raised some really interesting research questions that are worth exploring but I'd be interested if you would be willing to propose some research methods or some potential projects that might get at the specific aspects that you were talking about so a little bit more of a translation of the questions into a project or a methodology yeah for having thought for 20 seconds I'm going to go back to what I really emphasized in my presentation which is clinical decision support and I think that after 20 years of trying to encourage primary care doctors to adopt genetics into their scope of practice I realize that is just not feasible without these tools and so I think I would highly propose an implementation study around two to three different models of clinical decision support that both provide just in time information to the clinician and education to the clinician around how to interpret and integrate one specific screening which could be something like pharmacogenomics could be something like hereditary cancer testing and and really look at the success of that implementation and comparing those different products as well as strongly exploring the user experience from the clinician's standpoint in addition to that I think I really would encourage us to engage the community of patients and we're not hearing from them today specifically although I think all of us are also community and patients but we I think we really need to find out what this means to patients and how they would like to see this impact their care going forward so short answer Carol can I hop in here for a second we've got a fair amount of experience with providing clinical decision support to primary care providers we've been doing that for quite a few years both in the area pharmacogenomics and the other area of genetics genomics writ large and the experience that we have is that they don't want it so how do we get over that barrier so I don't know if you mean the experience in your specific institution in our specific institution but I would say more broadly I can report that at least I would say across the eMERGE network which has 8 or 10 clinical sites that's a fairly consistent result pharmacogenomics is a good example we provide pharmacogenomics clinical decision support and they don't want to pay attention to it they want to just click it off and ignore it so how do we get past that so I think that's where that end user experience is critical because I think we a lot of us designed that clinical decision support with interruptive alerts and primary care providers are overwhelmed by interruptive alerts and don't like them and so I think that implementation science rolling out an intervention that won't be adopted or received by the end users is going to always fail I think in addition to that historically primary care providers have not felt that genetics has yet reached the level of value for their patients and until we show the value of this of integrating this that it's worth their time then they're going to continue to think that this is more of a nuisance than a benefit yeah I really want to echo your second point there because the first point we all understand you know alert fatigue and what the impact of that is but our experience is the more fundamental problem is people don't even believe the evidence and so how do we get I mean when in fact I mean the experts believe yet so what do we need to do to fix that yeah I would weigh in and say that while I agree with you know what you are recommending I think we've got one step before that that we have to do and this came out of GM 13 so I'm referencing a prior and that happened to be on Informatics Research Agenda and one of the things that clearly emerged there was the idea that people really don't like an impositional model which is I am from at least I'm here to help you I'm going to solve all your problems and they say well we don't have any problems for you to solve thank you very much whereas if we actually sit down and say what are the things, what are your pain points and what are the issues and so I'll just use one example from our institution where there's an institution-wide initiative to improve colorectal cancer screening and so there's everybody's body and there's a big quality initiative to build a piece in this for those people that are at the highest risk and then let's sit down with our clinicians to say okay you tell us how you want us to do it and then when we do that we get actually much higher uptake because they're part of the design process and then of course we also follow up with the user testing afterwards because as we say every time when we roll something out we guarantee it's wrong or your money back the second piece is we'll fix it but I think we really need to think more about that early engagement in defining the problem and defining the solutions Carol I've been trying to say we've had a lot of discussions with our primary care providers about this and they're really not interested and I have to say and they're scared but our specialty providers are narrowly interested and they really want to do this and so we've really intentionally moved this and the things that we've done to mostly to specialty care providers and I just want to sort of contrast that because I really we've really found that offering our specialty providers are more and more themselves ordering genetic testing and offering them support around what they really want to do which is order genetic testing has been much more successful than our primary care providers who do not feel comfortable even with placing consults for genetics and so I just I'm pointing that out as some alternative models of thinking about how we would do it it's possible that a research topic could be contrasting in primary care versus doing it in specialty care and seeing what the uptake is because my impression would be the uptake and the interest is quite different. Thank you. I wanted to have space with you Kelly you brought up cascade testing a couple of times in your study and in our studies there is a lot of family communication and people are talking about their results that are negative with their family members but that doesn't translate to cascade testing and so I wanted to get your thoughts on how on how cascade testing fits into this idea of at some point we will be scaling up to large scale population screening but in the meantime as only some people are getting testing that is a great way to identify families at risk but there are huge barriers there and so is that part of what you think should be studied more in depth as we are getting to this larger scale population screening like is that going to be an interim approach? Yeah I think one of the one of the big barriers there is the fact that when you think about a family they are widely spaced out and even if you can communicate a risk to another family member the roadmap for that family member to go get the follow-up care and testing that they need there are hoops upon hoops that they potentially have to jump through to do that and some of that I think is thinking about how can we as in the disclosure in the communication to the provenance of the people that we are engaging with are there better ways to pave that path for those relatives and passing off resources that are not just for the patient but are actually for those other relatives and make that easier making it more clear and making it making sure that no matter where that relative lives that there is a way for them to get that care that they need and building those pathways and pipelines and communicating that through the patients themselves I do think that is an area right for research that if we can get that if we can improve that we can exponentially improve the impacts of population screening that by we can fix that pipeline and make that easier you can impact many more people by every person that we are identifying through population screening and I know there are some efforts out there through AI chat those types of things to help facilitate that and it's been really interesting research and I think that's a place where we need to do more and that we can do more testing of different models and figure out what works best Carol Jillian has a question go ahead and then Ned did you withdraw you go okay okay alright there we go and then after this we'll go to George this question I think is primarily for Melinda and Kali or maybe others in the room who know about this do you think of what you're doing in primary care under the umbrella of collaborative care models that are being done in psychiatry mental health with primary care and do you think that there are learnings from those models that we should be adopting because it does seem like they're solving some pretty important problems in this space including like even CBT codes for how those models can be sustainable so I'd be curious for your thoughts on that and also where those models may not apply in genetics I mean I think the short answer is yes you know I think that there's a lot to be learned there and thinking about thinking about what are all the other issues that come along with that in terms of liability and this and that and how those things play together and how we can build models that that are scalable that you also have to figure out how to have those things being reimbursed at some point that has to get paid for the effort of these providers in that more support role yeah so I agree I think that the collaborative care models we have a lot to learn from and I think just like you were saying is that the big take away from collaborative care is giving the primary care provider that confidence that safety net I think someone used that term maybe used that term earlier and I think that's exactly what I'm advocating for in replicating now in scalable tools that are creating that same degree of safety net and confidence in the tools that can allow us to scale even beyond but I don't think that I don't think it's wrong at all for us to be starting in this intermediary space with this collaborative care model and just like we have a pharmacist serving perhaps 20 different clinicians that we could have a genetic counselor serving 20 different clinicians as that resource so I think it truly is that concept of having that expertise and that confidence building to be able to move forward that we need to really be learning how to integrate and to scale so George and then April and then Carol thanks George Manson from NHLB but maybe to paraphrase Mark I should say George Manson I'm from the federal government and I'm getting help but you know the challenge you described extends beyond primary care I see the same thing in family medicine and the comment I want to make is would capture that as one of the critical challenges that's really right for research so for example the National Academy of Medicine conceptual model for meaningful community engagement rather than working separately and then bringing the answer to here I have a model to help you it's not going to work but it definitely would work if we re-approach it by using a similar conceptual approach of meaningful engagement with whether it's primary care or family medicine and really developing the solutions that it's co-created that's more likely to work I think community engagement and just a teaser for the next session coming up and implementation research can really be very helpful in addressing it thank you April I just wanted to mention when we were talking about engaging primary care and how it's easier with sub-specialists and things like that and I think that we have to really be careful about focusing a lot on sub-specialists because that's where a lot of people have bigger barriers people have bigger barriers getting into sub-specialty clinics and so really thinking about thinking about the long-term thing so what are we doing in our medical schools in our residency programs and all of those things to make sure that people who are coming through understand how genetics is incorporated into whatever specialty they're a part of and having that being a more robust ongoing education so that we are not losing that interval time period to say okay even though you're going to be a family practice physician you should still know these are the top-down levels of where genetics is going to be a part of your patient's care so it's not an ad for them it's just a part of their regular flow the same way we order a check-sex rate to evaluate for pneumonia or whatever thank you Carol Horowitz yeah um thank you and thank you for what you said George I completely agree with you I you know we we do our work in many academic community and safety net family medicine and primary care practices we're and I I'd be I'm interested in learning from you what the problem is we do what George said which is we bring those frontline clinicians in to develop everything with us every step of the way they're saying this will work this won't work for me the message they give us is we care about this we just don't care about as much as you all do because we have 50 other things to care about at the same time and we balance it out so I'm interested from you are you finding that people are not interested or that they're balancing primary care people Carol you you cut out there just briefly at the very end could you restate that I was asking if you're finding that the primary care folks clinicians are not interested or are they just balancing with everything else on their plates would anyone like to respond to that Rex maybe yeah I can I can at least tell you what our experience is and you know I think it's it's both of those things you know they're very busy they have what is it seven minutes per patient that they need to get them through so that's that's a problem and then the increased demands I think of messaging through the EHR and dealing with all of that has put additional stress on people in the primary care community so I want to acknowledge up front that they've got a lot on their plates already but what I what I'm struggling with is the disconnect between people saying the primary care community should be doing more of this with the experience that we've had which is they don't have to fit it in and and they and I think several people have said this they they're maybe a little nervous about making a mistake because they're not well enough prepared for this but then I think all of it the second piece of that though which does surprise me a lot more is and maybe it's unusual at our place because we're an academic health center right we're not a we're not a community-based organization and what we've experienced is they take initiative they go out and they read the papers and they're unpersuited by the literature that persuade it's the rest of us and that's a separate problem I don't know how to overcome and you know is it just that we're true believers and since we're true believers we're willing to accept it or is it that we still need to the idea of a research agenda I think this was off the table though but that we should we need to generate better and more convincing and compelling evidence that what we're proposing is of value is it ignorance or apathy I don't know and I don't care so I was just going to add a few comments there so I think there's a couple things one I think there's a difference between thinking about overall integrating genetic testing beyond genetic specialists and I totally agree that the subspecialists are highly engaged and highly motivated to do that because there's so much that's impacting their ability to actually do diagnosis now and genetic guided treatment for those patients versus screening right and population-based screening and really the population-based screening being in that primary care wheelhouse but one of the things that you know we need to do to make I agree that the providers are interested but busy and overwhelmed but we're not helping them prioritize this because there are no guidelines right their societies don't have any guidelines you know until it becomes a checkbox that the USPSTF requires primary care doctors to think about that's really hard for them to think about and then there's also no incentive and so we are paid to check other boxes and unfortunately that's the reality of how medicine is practiced and so we really need to think about not only proving again getting back to that value for their patients which I think primary care providers really care about but then value to themselves and guidelines for which to follow to do all of this so Jessica did you have something else and then at the end of the table here I think and Jonathan and then Jonathan next to you oh Jonathan how do I miss you and then Dan so I just wanted to say we did a study within Kaiser Permanente of who was helping patients at high risk due to genetic variants with their follow-up care and everyone pointed to the other primary care pointed to the specialist, the specialist pointed to primary care genetics was saying we would like to do it but they're not coming back to us they're getting their diagnosis and they're leaving and doing chart reviews a lot of times the guidelines in their record are outdated because there's no contact to help them update it and so I think it's a problem on both ends and who do we expect to be taking care of them and the follow-up and is that who we can guide to get the testing in the first place can it all be grouped together as part of care and so anyway I just wanted to emphasize that's even in a well-resourced system of everyone was pointing at the other and at the end of the day the patient was saying I'm taking care of myself because no one's helping me do it and I don't think it was for lack of interest I think it was they generally thought someone else was doing it at the end of the table here hi okay this is not going to be me Ilana at Geisinger and I want to echo also what Jessica just said you know we when I did this in the collaborative model with psychology co-located in primary care when I was at Kaiser we actually experienced sort of the same thing in our co-location collaborative model ended up with a visible invisible visible wall in between psychiatry and primary care how that happened even though they shared a waiting room I have no idea but it was amazing so but I also wanted to get so that is one issue and I wanted to get also at what George and Carol were saying too though is you know we've heard this co-creation co the engagement in the co-creation and how important that is and it sounds like there is some movement in it working and I wanted to hear what has been created then and so you know we keep talking about what we're doing so far isn't working so we need to co-create things that could work so what have we found so far that may be working or maybe those first steps on the pathway and maybe you haven't gotten that far yet with the after you've co-created it you haven't actually tested it out yet and that's okay but I'm wondering if there is anything on what is working so far when you do that co-creation comments from any of our speakers to that point or anybody else in the room so when we were setting things up for our screening program the question was does it sit in a clinical situation and go through clinical care or does it become a research protocol and through the co-creation and discussions with primary care specialists and others they wanted it as a research protocol and so we pulled it out of what would be clinical care and made it into an IRB protocol it's a whole separate team it's ordered by one physician that is the PI of NRD and so that's not maybe answering the exact question Alana about primary care but that was a community engaged approach where ultimately the decision was that it's not going to go into the clinical pathway it's going to go in the research world. And did you in that co-creation process do you have the why they wanted it as a research? I think a lot of it is what we've been talking about just being concerned about feeling comfortable ordering the volume our physicians are happy to talk about the program but then to take it to the next step of actually being responsible for some of the ordering and follow-up they wanted that to be really outside of what what they're doing on the day today so we've exported that to our research protocol so it wasn't about that we need more research it was about that they was practical and needed somebody else to take on the burden. Yep, exactly. So we're getting pretty close to the end of time so Kelly if you want to make a comment and then we'll do three quick questions and then I think we'll be pretty close to the end. Sure, I was just going to echo that we've had very similar feedback and it comes down to logistics of just in talking with our providers from an implementation standpoint with us working with primary care clinics in Alabama that the providers see a lot of value and benefit in having people there that are managing this project that are they do that nurses and people that are part of the research team doing that navigation that can do this all day long and get really good at this talking points and talking to patients about it and kind of having that as a unique role that's kind of sitting in that and it's not falling to the providers that is afforded by a research study though and so at some point you've got to figure out how that can become part of clinical care but thinking about who's doing that work and whether there's a role for somebody doing some of that effort specifically. John? So my comment is on something a little bit different it's been a great discussion but April there was something that you said about drawing lessons from ancestry based carrier screening and I think we could probably all agree that we want to avoid down the road of ancestry specific screening programs but wonder if you would comment on whether you think that including there's we certainly want to include conditions that are of importance to particular groups so thinking about April 1 for example is that something that we could envision as a vehicle for community engagement around a condition that is of importance to that community but also then broaden the opportunity for screening for the other conditions that are important for population screening. Thanks. So your question is really like if we take this condition that's known to impact this population and then use that to start the conversation for what's next right and I do think that that is you know a really a good point so like for example if you take sickle cell right so you take sickle cell disease which impacts African-American communities and there's a big disconnect between I have diagnosis and I get treatment and what those barriers are and so talking about what we know we can screen for this and identify you and these are the strides we're making to improve your health via this and showing you're building that trust in a system that maybe we forgot about this here but we're coming back and we're including this population now and so building that trust that's definitely gives you that opportunity to take the next step so I have definitely agree that's a very good thought process and how can you you know incorporate some of those things but I think the other piece of that though is you definitely have to have somebody who's in the community who's also already has that trust right so you have to kind of have multiple layers of building trust with people who maybe don't trust the system that exists. Dan. Despite even changing the subject or sort of opening up a can of words but I'd like comments from any of the speakers about the mechanics of cascade screening so the issue is you have to rely on the family member to get the other family members to do that there's a real communications problem there both in terms of what they tell their family members and whether they tell their family members understand or respond and I think it's sort of a gap in the way we deliver genomic care and I'd be interested in hearing how you think it how you deal with it and how you think it ought to be dealt with. Yeah that's a great point you know I think that evidence from a number of studies is that families I mean there are absolute exceptions to this rule and lots of reasons some families don't communicate but largely when we ask people do you talk to about your results or that you had this testing the overwhelming majority of people are talking to at least some of their relatives whether those are the same relatives as the ones that would be at risk I don't know but I think we have an opportunity the spouses are unrelated to them usually at least in except in Tennessee and Alabama maybe yeah right right but I think that there is it's not a perfect solution that there's an opportunity to help improve that normal communication that is happening but thinking about how not to rely on a patient describing those results correctly or what they should do with them and additional tools and resources that can be built and handed off where that all you're relying on the patient to do is to pass this to this other person and that there's enough information there for them to not only in their risk but to know what on earth to do with that risk if they live in Alabama or California or wherever else but I think it's also important to note there's been two or three studies now where direct contact has been used where they haven't gone through a familial intermediary other than initial permission and the cascade testing rate is exactly the same as familial communication so there's something beyond just communicating within families that is depressing the uptake of the cascade testing Contact to family members permitted with permission from the family Yes, so that's allowable with permission from the program in every state as well right so I mean you have to consider like do those patients have access to a lab do they have access to pay for this testing so every barrier you bring down for that initial contact has to be done for all those family members also and then also you have to consider that some people there may be stigma associated with having an abnormal test result all of those things and so even if they've told their family maybe it stopped there because they're like we don't want to talk about this and so I think yeah there's probably a lot of other barriers as you go down that cascade that are going to continue to pop up This is one of the value ads of population screening though right we actually take that burden away from families you know if we think of newborn screening we don't do we don't really do cascade testing on newborn screening results because we have confidence in the system and that everyone is accessing accessing their their screening Yeah it becomes a transient problem in a future world Dan one more comment No So I think we're up to time here so just quickly on this top of the logistics this was a far-ranging discussion but you know earlier we had a comment on logistics by Christine about sample collection tracking and all that stuff that is an important part of the logistics that I think we're talking about here and in this session covered a wide range of topics who to test when to test who does the testing how to report who reports who takes primary responsibility for the care is it primary care sub specialties preparing the workforce the timeliness how it's delivered when it's delivered the needs still for the informatics infrastructure to align to make that information exchange seamless I think we could talk for days on that and I would love to sort of informing the public conforming physicians informing families rolling this out in an iterative fashion versus trying to do it all at once we talked about risk estimates and the idea of maybe systematic or automated processes for implementing guidelines on who to test rather than having very complex guidelines that are difficult to actually manage and the idea that a combination of approaches population wide and those focused on high risk groups will be needed to do this effectively we talked a lot about sort of the challenges of underestimating the diversity in patient populations and not capturing that diversity adequately so the multi-dimensional nature of identity that crosses race and ethnicity and ancestry we don't represent that very well and it is an important part of the logistics of actually delivering genomic medicine at the population scale and there were a couple other points in there the barriers to access the cost and everything is not equally shared across the population there is logistics around circumventing provider bias and discrimination and then finally understanding provider training and patient education and communicating especially I thought from that an accurate picture of risk based on screening so that people don't underestimate their risk or overestimate their risk and I think that balance of that is definitely one of the logistical things that we need to tackle and there's probably great research project around that so that's my quick summary of the logistics Super thank you Carol so at this point we have a break we'll break until 325 Eastern and then the back for our final session today which is on community engagement which has come up quite a lot so see you at 325 and thanks to all our speakers