 In September of last year, NHGRI held a roundtable here in Rockville, and the subject was addressing the challenges associated with diversity, bringing diversity samples to genomics research. Vince, who is the senior advisor to the NHGRI director on genomics and health disparities, was the organizer of that, and there was a summary presented in the ECB. Vince is going to give you a report on the roundtable and ask for some feedback from Council about strategies for, again, increasing participation of underrepresented groups in genomics. So good afternoon. I'm really pleased to have this opportunity this afternoon to come and to brief you on the roundtable that we had last September, but also to share with you some of the activities that are going on within the Institute related to genomics and health disparities. As I listen to Jen and I know Lucia coming after me this afternoon, that really this is a thread of a conversation about how we are thinking as an institute about the issues of underrepresented populations in genomics research and the questions of disparities in the role that our institute and genomic science and LC research has in understanding health inequities and health disparities in our country. So I'll take you through very briefly some of the activities going on within our institute around genomics and health disparities and then talk to you about the recommendations from the roundtable and really look to the Council to have a conversation about priorities and opportunities for NHGRI. So as stated by Rudy, I shifted roles this last year and one of my major responsibilities is to advise Dr. Green and the senior leadership around issues of genomics and health disparities and to help us in moving forward the conversations and thinking about the research portfolio and what we should be doing in this area. And part of that is actually through data collection, collecting data with regards to the various portfolios that we have with regards to research today, both in the intramural program and in the extramural program to use for strategic planning and to help to guide senior leadership in thinking about these issues and the responsibility of our institute. We also have an obligation as part of NIH to report on what we are doing around minority health and health disparities and to lead those efforts for our institute in reporting at the NIH-wide level. But one of the, I think one of the most important parts of my job is talking to other institutes and other federal agencies and seeking to build stronger collaborations around health disparities, health inequities research related to genomics with other institutes and other parts of the government. And finally to actually do some education outreach efforts here at NIH as we help to support and facilitate these activities throughout the scientific community. One of the efforts that we're doing here at the National Institutes of Health in collaboration with the National Institute of Minority Health and Health Disparities, NHLBI, NIDDK, the FDA Office of Minority Health is to sponsor a lecture series on genomics and health disparities. This is primarily for staff here at NIH and the trainees here at NIH, but it is videotaped and will be online so that it can be used by the scientific community. And you see here we have three lectures coming up this 2016 year and we are now planning for 2017. We would love advice from Council about potential speakers and dealing with a broad range of topic areas and disciplines of importance and relevance to genomics and health disparities. And again, this will be online and be available both in teaching settings that you may have as well as to be used within other types of areas of importance to your work. So my primary focus here today is to talk about the roundtable. And the roundtable occurred on September 16th, as stated. And I want to start by just acknowledging that we had two Council members in attendance, Dr. Halbert Hughes Halbert and Dr. Boerwinkle, as well as a number of extramural experts that we brought in to participate in that meeting. But we also had many program staff from across the institute involved in the meeting, attending, observing, watching, and we had six other institutes attending and participating in the meeting. Again, this building stronger collaborations across NIH around the fields and questions of importance, of inclusion of underrepresented populations in genomics research. So what were the roundtable's objectives? The first objective was to bring together genomic scientists and health disparities researchers who actively are involved in working with populations that are underrepresented in genomics. So we really wanted to bring people from different disciplines but have been involved in actual research and activities so that they could bring their voice and perspective from their work, but also to have the opportunity to have this kind of really exchange from different disciplinary scientific perspectives. And to discuss the scientific problems brought about by the lack of diverse ancestral populations in genomics research. I want to stop for a second and say that this roundtable was focused on that question. How do we increase the diversity of ancestral populations in genomics research? We're talking today about both issues of disparity populations, but we're also talking about when we think about more broadly, genomic science, the need to have more diverse ancestral populations. And we just heard from Jen about some of the exciting things that have happened with H3Africa. And so that was the focus of the meeting. We all know the problem that we have, that the primarily the populations in our research to date have been European populations. And how do we shift that to really have it reflective of the world and the genetic variation that we know exists across the globe? The third objective was to identify existing barriers that limit underrepresented populations from participating in genomics research and strategies to address those barriers. So it was to identify both barriers and strategies for success. And then finally to identify opportunities to study health disparities and health inequities in genomic science and LC research. So now let me just frame and provide you the recommendations that came from the participants in the roundtable. I'm going to walk through these. There are five areas. But I'm really looking forward to the conversation we have after I present the recommendations. So the first area is for NHGRI to provide leadership for the field of genomics in promoting health disparities research using genomic methods. This was a broad theme that was presented by the participants in the roundtable that we were in a unique place to really take leadership in helping to move forward both what happens here at NIH more broadly in the genomic science community. So specific recommendations related to that was support the formation of interdisciplinary centers of excellence in genomics and health disparities. And this potentially could be in collaboration with other institutes. In genomics research they incorporate social and environmental exposures. This was a major theme about the importance of studying in a robust way social and environmental measures, conduct genomic studies of underrepresented populations that are inclusive of translational and implementation research, and convene conversations and meetings to explore opportunities to collaborate in areas of genomics and health disparities. Second area, examine institutional factors that impede greater inclusion of underrepresented populations in genomics research. And this issue of institutional factors really came up as an issue that there are times when because of institutional factors we're not able to have the best research funded and the need to address that issue. And so one area was to assure that study sections have the expertise to evaluate applications studying underrepresented populations in genomics research so that we're making sure that the members of the study section have the ability to really analyze the applications that they're reviewing, enhance review criteria to evaluate inclusion of underrepresented populations in the application review process. One of the issues was raised at the timing in the review that it's really part of the secondary review to make sure that certain things are addressed by the application versus the study of the scientific merit and recognize the greater exposure of longer term recruitment timelines and for certain underrepresented populations. This issue that sometimes it takes longer to recruit, building the relationships with communities, the cost may be different and that that's an important issue to recognize upfront. The third area is to promote the incorporation of environmental data into genomic studies. Again, that theme was a major theme of incorporating environmental variables and strategies to come up with that. So this question about how do we look at multifractorial determinants of health and how is that really integrated within the genomics research that our institute funds and we collaborate with other institutes. Fourth area, to promote comprehensive strategies for analysis of existing biospecimens from underrepresented populations to incentivize processing and analysis of existing sample collections from the ancestrally underrepresented populations. The concern was that, you know, there's been a lot of recruitment and there are samples out there. Should we as an institute and should we with other institutes seek to get people to go back and really focus in and target on those populations that are already in the data sharing available genomic different websites that we have access to the data sequencing data. Support the creation of new analysis methods and localized centers of expertise to analyze those samples. This was an issue about that the need for methods, that we have some unique needs for methods that we need to think about diverse ancestral populations and this was articulated in a way that there's really this opportunity and need to be able to understand and to do the analysis. And so methods development is also an important area of focus. And to lead the development of a national genomics resource of diverse populations so that we take on some responsibility to say that we actually need to very much have diverse population samples that are available for the broader community. The final area of recommendations was to support sustained community engagement and community-based participatory research, to stimulate dialogues, to hear community perspectives, that this is an important component of the ability to work with diverse populations within the United States. To build long-term relationships with communities, including capacity to receive and respond to community feedback. This question of how do you build a long-term, ongoing relationship with communities? How can we actually as an institute to help to do this with more within genomics research? Surely it's an area that has been of importance to our ELSI program, but can we expand what we're doing around community-based research and study recruitment and recruitment methods in new and existing grants in the portfolio. So this is getting back to the issue of strategies to address the problem of lack of diversity and diverse ancestral populations that if we study recruitment issues as part of our portfolio of research. So in summary, I want to just highlight some of the recommendations that I just articulated that were common themes that we heard during the roundtable. NHGRI needs to provide leadership in framing disparities research, including baseline commitment to diversity of cohort participants and reduce cohorts of convenience. Application review criteria should include formal scientific consideration of inclusion of underrepresented groups, support the creation of methods to analyze diverse samples, and to recognize the need for flexible timelines needed to engage diverse populations. So we're moving forward. I think you're hearing and you're seeing that we're moving forward in this area, but we really want your input and we want your specific advice of opportunities to prioritize, recognizing that there's limited resources and we're not going to be able to do all the things that we could do to actually enhance the inclusion of underrepresented populations as well as issues around more inequity and disparities research. So we know that we can't do everything. So where's the greatest impact? Where are the opportunities? What would you recommend that we prioritize? We want you to help us to be more specific starting today with that conversation but coming back to you and in different forums to really explore these issues, but we really want guidance to help us and you have everyone in the room kind of listening to this. And so we look forward to this conversation. I would like to ask Dr. Hughes-Halbert and Dr. Bourwinkle to start out since they were at the round table, but I look forward to the conversation. Thanks, Vince. That was a great presentation and I would just like to take this opportunity to congratulate you and the round table co-organizers for sponsoring and holding a wonderful and insightful session. It was truly a great exchange among individuals with the diverse types of experience and expertise as it relates to disparities and genomics. So I had a couple of thoughts about priorities and the recommendations I think were really great and are very succinct and sort of straightforward. The things that resonated with me the most, just thinking about the tenor of the conversation was that many of the barriers that were described by folks at the round table are long-standing barriers. I think we know about those. There's been a lot of data published about why people will and will not, people from diverse groups will and will not participate in genomic research and sort of the issue about these cohorts of convenience really limiting our understanding of the ways in which genomic factors contribute to disease risk and outcomes in ancestral populations, which actually like that term. So I would agree with you that I think NHGRI has an opportunity and as I see it, a lot of the, I can see the effects of the round table sort of having an impact now think about the CSER, new concepts that are being put forth. I think of sort of taking these recommendations to heart by thinking about issues related to diversity and inclusion in a very real and concrete and substantive way. You know, I think, so I had a priority list, I think the Interdisciplinary Centers for Genomic and Disparities is really an important place to start. And I was going to say that would be my top priority, but as I was listening to your comments about the limited resources, it strikes me that there's a wealth of knowledge and data that perhaps haven't been mined sufficiently to understand some of the, to address some of the issues that you've talked about. And so thinking about my work, Gail's work, and here has some, I think some, touches disparities in some meaningful way. And it might be worth trying to figure out how to mine that data to identify ways to address some of the issues that you raise. And as I'm talking, I'm sort of losing my train of thought because of this one. But that's sort of maybe where I would start in terms of priorities, sort of leveraging existing data and existing resources to really try to think about, well, how can we engage populations from the perspective of identifying what their concerns and priorities are? I would think that there's been a lot of work that has been done to understand barriers and facilitators to participating in genomics research. And if you think about, well, we wouldn't, if one thing that's consistent across studies might be that folks aren't interested in the disease, they don't understand genomics. And so that could be a way of sort of thinking about community priorities and identifying community priorities without sort of funding a special initiative. So that might be an important place to start. I do think that, I'm going to stop there. I'm going to give someone else an opportunity to talk. Since first thanks for your leadership and pulling this together, I certainly appreciate it and many other people do. I really like the meeting and the outcome for, I'd say, three reasons. One is we didn't get drawn in, as we often do, into sort of cliches and platitudes. We really stuck, I think, to the mission of what we're going to do about it. And I thought that was very important for the outcome of this meeting. Then second is it, you know, one of those cliches that we get drawn into very quickly is sort of political correctness in this area. And what we did is I think we outlined good sound scientific reasons to promote diversity in large scale genomic research. And the two that come to mind are just the site frequency spectrum is very different among ancestral populations. So those that ignore diversity are really going to do it at their own peril because many of these rare variants are population specific. And there's an element of chance in this area that as scientists probably makes us a little queasy. But one way to remove that element of chance is to have diversity in our samples. And the second is gene environment interaction. You know, we give so much lip service to gene environment interaction, but very few people roll up their sleeves and try to attempt to address it in a translational setting. And that's, again, a reason to the meeting I thought was special. And the third and a related reason is we did not limit ourselves, in fact maybe the pendulum swung a little too far in our sort of euphoria. We didn't limit ourselves to skin color and ancestral diversity. We spent a lot of time thinking about the whole spectrum of diversity that influences health and disease, economic diversity, the social setting, on and on and on, and the impact of that. And I thought that was a very important outcome for this meeting. I didn't, sorry, that turned into a speech. But then coming back to sort of your list of recommendations, I probably would still stay with the specialized centers to promote, even though it would cost monies to do so. But I think to address some of the scientific issues, I really think it's going to, you know, they can be virtual centers, if you will, speaking of cliches. But they can be, you know, they don't need a single geographic location, but a coordinated network of individuals addressing some of the scientific challenges that you brought forward. And I think that would work and make a little bit of progress in this area. And I said a little bit of progress on purpose, because going back, I think NHGRI is in a, not a unique, but a special position that really, I think we can actually, you know, serve as a hub or as a core for the other institutes in in promoting scientific ways of addressing health disparities through inclusion of diverse populations and peoples. Thank you. Yeah, so Aaron, you spoke earlier today about $130 million for the precision medicine initiative that will collect samples of chronic health records in physical exam measurements. And my understanding is by July of 2017 there will be 100,000 of those available. So I would just think that it would make more sense to think prospectively to do that than to go back to old collections. So this doesn't get as comprehensive as what Eric just described for, you know, taking a look at diversity from a more comprehensive perspective, but I was struck by the previous speaker and I just wonder whether maybe dealing with at least a part of the underrepresented populations, the ancestral populations of Africa that the H3Africa program could be folded into this initiative to at least address part of the underrepresented populations in this country with the African Americans. I guess my only reaction and comment I would make is that going back to this issue of data collection and as we monitor kind of what is the ancestral diversity of the populations that are available in the open databases, that's something that we clearly can do and that will help us understand that. So as a recognition of that question of kind of are we enhancing the inclusion of those populations that are left out? So there's synergy I guess is what I'm saying. Right, and to clarify I mean the H3Africa program is not ours, it's the common funds we lead on, we donate money to it, but you know it's not ours to amalgamate with one of ours, but certainly to synergize with it, et cetera, et cetera. Yeah, and leverage those resources. Exactly. Yeah. All right, Gail and Jim, and then she'll be there. Yeah, well then you know, we just saw each other at another meeting that was out with a lot of these same topics and I think this is a very good presentation, very good to spark conversation. I want to just say first I really support what Janita said about the fact that there's a ton of data a ton of information already about the ways that social and environmental factors that are related to race and ethnicity and other vulnerable populations identified through socioeconomic status, through, you know, remote geographic areas, et cetera. We know a lot about barriers to access to health care and we know a lot about the ways that you know, that's sort of the causes of health disparities in the United States that are very linked to treatment of individuals through racism, through and mistrust of research. I mean, we have done that, okay. We've also really had the conversation about the fact that the OMB categories for race that we all have to report and ethnicity that we all have to report whenever we get a grant. Those are very inaccurate measures of biological phenomenon, but much more socially constructed. We've had that conversation. We know that. So one thing that I really want to strongly recommend here is that we work very hard to not mix up a discussion of disparities with a discussion of the contribution of ancestral populations to health outcomes, to risk for disease, to well, to risk for disease. So I also really like having a new term because the minute we start using terms for diverse populations that we're using for other reasons and other contexts to actually represent other determinants, we get lost. We just always do. And then we dig our way out and we say, oh, well, really, you know, we know it's very complicated. We know these things are measuring different things. But what are we going to use for ancestral populations? We often use race and ethnicity because if we're going to do a survey or we're going to do recruitment, how do we decide who to include? Right? We don't have their genetic data already to sample from. So what do we do? And then so we sort of fall back on that and we hope for the best. But then we're already mixing things up. It's already not the best way to recruit people if we're thinking about ancestry populations and it muddies the water about causation. So I would say that a wonderful, this is a wonderful opportunity for NSGRI to take a lead in really thinking about what is a good way to measure ancestry populations to be a proxy for it in some way that when you say you're going to do inclusion of people that need to be included in genomics not in environment. Environment, you know, we can think a whole lot about factors that are related to environment like exposures that are related to socioeconomic status that are related to discrimination. But this is only about genomics. If we really focus on genomics then we just have this amazing opportunity to say we haven't been thinking about this clearly it really needs to be differentiated it's really tough to come up with a way to do this. In fact it's probably impossible we probably need to do it in ongoing studies where we recruit people and then we test them and then we say well how close are we? If we use some proxy how close are we? Well we're not very close. Well then how could we get closer? Anyway this is sort of a soapbox that I'm on but I think that it's just a moment where we can say we're not thinking about this clearly we really need to think about it more clearly if we're going to talk about inclusion. Okay, thank you. Great. So we were all charged with trying to come up with some very specific types of actions that we could make to increase minority participation. So just trying to run down a list of the things that I thought would be most helpful. One is that there I love your presentation I wish I could have been at the meeting. One of your bullets said study recruitment and recruitment methods. As I mentioned in closed session I think one of the really critical things we have to do is not just study recruitment we have to study the whole cascade and retention and although that sounds kind of platitudinous it really stunned me when we looked at our own enrollment cascade. We've done pretty well in our CSER project of really targeting minority involvement but if you look at the enrollment cascade what you see is a disproportionate loss of African Americans for example and as Gail I said this in the closed session but I think it's really important as Gail has said to me if you look at this enrollment cascade you end up you select these people and then you get a subset of those people and a subset of those people you've got a tiny subset of people and you're trying to make general conclusions and that's kind of both. So to cut to the chase I think it's critical one of the things that HGRI can do is to demand some reflection of what that retention and study cascade looks like in studies. That isn't particularly expensive. I think that one has to ask for very detailed and concrete plans of how grantees are going to accomplish these things and we have found that working with clinics that have high minority populations funding transportation integrating with the subjects clinical activities because they're much more likely to go to their clinic visits than they are to an add-on research type of visit. Having language resources obtaining backup contact information. These are populations where you can't get a hold of people at a disproportionate rate. So those are fairly simple things that can be asked about. I think that you can ask grantees to lay out a plan for how they're going to engage the community and the subjects because we've found that as it's not new to anybody who's looked at the literature on this trust is an issue. So besides very tangible things like I can't afford gas to get there and we really see an increase in enrollment if you give gas money trying to find ways to increase trust although more difficult is still doable engaging the community newsletters working with clinics that have already gained their trust that they take care of them. I think really briefly that issue of how to define the minorities is really important. Again would urge you to go back again and again to groups that have had disparities and I agree with Gail you can't conflate right disparities with this but you do need to use that as the touchstone I think for deciding what groups you're going to target otherwise it becomes kind of an exercise in nonsense. And as far as highest priority I guess I would just say focus on strategies for looking at the entire lifetime of the study and how grantees do with retaining individuals not just recruiting so I'll stop there. So listening to the discussion made me think of a question about this idea of an interdisciplinary center for genomics and disparities which in some ways I would think of that as being implemented as one-time solicitation for the established centers versus thinking more programmatically in terms of creating a program in HGRI that focused on the issue of disparities in genomics and then that program provides the infrastructure for addressing all the recommendations that was developed through this workshop because it was broader than inclusion I think. Patient engagement is important but it's not just important for getting to recruiting and retaining individuals in studies it's an issue about defining the research agenda and so I don't know the rules and regulations for creating another programmatic entity within an NIH center but that may be one option to think about because it would give the infrastructure to address all of these issues in a empirical scientific way rather than trying to figure out how to integrate as the only strategy. I think integration is important in sort of having a broad impact but I also think if it's too diffused and not coordinated in a sufficient way then it becomes too diffused and meaningless and you can't see its impact so I don't know what the I guess I have a question in a comment and now my question is, is it possible to create a different organizational program or I guess what do you mean by organizational program? You mean another program in the extramural part of our portfolio? Yes. Well we do that all the time. We could in principle need to be budget for it but sure. That's not something that we do within the government I mean you're saying organizational so I'm just wondering if you don't mean... In my mind organization is NHGRI so what I'm thinking about my parallel is in a cancer center there are different programs that address different scientific issues and the way I sort of make that relationship to NHGRI is that there's genomics to society there's the other programs so I might be conflating all of these... When you say problem do you mean sign that we'd be doing with government staff or do you mean a program we'd be a granting program we couple together of some sort of funds to do something where we'd be... I think a granting program is what I'm thinking of. Val and then Lon. Yeah I just wanted to follow up on that David's comment earlier about it makes sense to go back to the ancestral sites if you want ancestral populations regarding H3 Africa however you have to be careful there because one thing that H3 Africa and other international groups I've been involved with they don't want you exploiting their populations they want to do the work within their own countries and that's become much more prevalent now than it used to be 20 years ago when I started working outside populations so I don't know if it can be done or not then the other thing I want to comment on is very interested in Eric's great comments about diversity including environmental diversity I just wondered how much that was really discussed at this meeting coming from Iowa there's two things I can't get away from one is presidential politics yeah done the other is farming so we do have a lot of rural populations that have sort of unique exposures or somewhat unique exposures and I think you have to balance going to an environmental different population going with ancestral this would hurt me research wise but I think the emphasis would tend to be should be balanced towards the side of ancestral yeah maybe it's more of a question of clarification for you but also Eric because it's kind of in your area my question is and it relates something you had specifically on your slides but also that I think Gail and Jim came to and it's this analysis methods of reuse of the data that we've got it makes perfect sense to extract every ounce of information we can from the data that have been collected because they're so difficult to collect and costly and so on on the other hand there's an old adage this paraphrase that to ask someone to do and construct analytical methods after the data are collected is to do a postmortem and so is it how much can we get from the data that they're right data like Dr. Parker well first I don't see them as mutually exclusive I think first we should make sure we're analyzing slash mining the data we have with this discussion in mind typically people have other questions they're looking at the data they're not looking at the data with a particular eye in terms of either health disparities number one or issues of recruitment and retention there's a lot of experience in the epidemiologic community on recruitment and retention of different groups of people most people to be honest with you don't see it as science and so a lot of it's probably not published you know the stories that Jim just talked about there's just a lot of experience out there and we probably need to get it up and get it out and then the second you're absolutely right you know there's probably nothing worse than you know sort of backing into the question so if we can articulate a clear agenda I do think it's important that we do launch new initiatives in this area so again I'm repeating myself I don't see the two as mutually exclusive and they both should be undertaken and I guess the one comment I want to make is that one of the issues that came out during the meeting was this around methods and thinking about the issues of variants that are not known in certain population are there new methods that we could develop to better utilize and take care of the samples we already have and you know the person I'm remembering who was talking about this was a basic scientist who was thinking about this question of that we need to have some different approaches to address these issues so just to the last question under the discussion point what would we like NHGRA reports on efforts to include underrepresented populations so I think in the past we have had such reports about the success of recruiting underrepresented groups in genomics you get an annual summary but that's just a snapshot of what's going on it's not targeted in any way yeah so I mean I think in addition to that I don't think that report was asked Jim's issue about retention it was really who was recruited but not about the attrition rates and I think that would be really good to include in these reports maybe even along with the information on the kinds of things like our gas cards provided our things so that we can really start to break down what factors are the most impactful in retaining the recruits in the study I think that would be useful I don't think we've had that kind of report so I would like to see that and then just the second quick point is that again to something we talked about in the closed session I think one of the best ways to continue to be more successful in recruiting is to have the genomics research community also be more diverse and we have a lot of ways to go in that realm and certainly the diversity and inclusion plan all those training programs we have to continually push and maybe even expand the efforts in that because I think that in the long run will have a huge impact on our ability to engage these different communities well thank you very much I just want to close by saying that there were a lot of people involved in helping to make the round table a success but I really want to identify the individuals on this slide the staff and the genome institute that were involved and listened and were part of the process of helping to identify the individuals to invite to the meeting so I just want to thank my colleagues at the institute and my colleagues at other institutes that participated in the meeting thank you