 So welcome back to Family History. We had a great foundational talk this morning from Marin Schooner, and we had a very lively breakout discussion yesterday on what was thought to be the sunny terrace room, but it wasn't quite that sunny. We had a very diverse and a group of experts that both may not all be listed here, and I apologize for any misspellings of anybody's names. And we essentially, over the course of the afternoon, arrived at six different topics or potential project areas for family history, validation of the information that's obtained in the course of obtaining a family history, the implementation science to integrate family history into the clinical workflow, to develop the user interface for both patients and providers for family history, to develop the outcomes research agenda that would enhance uptake and adoption and provide evidence for clinical utility. We discussed briefly a holy grail of integration of all data and the development of disease risk models. And lastly, to assemble a group to facilitate the incorporation of family history initiatives into large cohort studies. So let me just go through these in a little bit more detail. So in terms of validation of family history, probably some of the more age old questions for family history is how do you know that it's accurate? And particularly patient entered data has the accuracy of that can be questioned. And there's also the question of how often one should be obtaining family history as pedigrees, as families, and our lives evolve. How does family history data also evolve with it? And what's the frequency that it needs to be obtained? And we thought that this could be the basis for some very focused initiatives in small projects in practice environments that are currently doing these or others to develop sets of iterative questions that could be used in follow-up of an initial family history and to refine those questions to obtain the most valuable information. And also perhaps a core leverage to that is to develop adaptive patient questionnaires that would move them through some logic to obtain accurate family history information. Now, some of these already may have been developed or are in development, but I think we wanted to leverage from those and move forward. This is also discussed in the context of David Valley's program in Mendelian disease genetics, where it's absolutely essential to obtain adequate validated family history information. And OMIM is being used as the entry point for a lot of the ascertainment for that project. And so there's an opportunity to collaborate with that ongoing research that we talked briefly about how would we benchmark validity and, of course, obtaining information from living affected or other family members would be quite important. So to compare patient, single patient-oriented information with those obtained from the family more broadly would be important. And a side note was we recognize that there may be a myriad of cultural and ethnic differences that might prohibit one from actually divulging important family history information. The example that was used in our discussion yesterday was about psychiatric illnesses and how patients might be willing or not to talk about that as well as, I think, you can imagine other disorders that we may not understand best, how to elicit from a cultural framework. So the second topic was the best way to integrate family history into the clinical workflow. And at least one of the participants in our working group had done some analysis of their electronic medical records showing that only 1% had adequate family history data, and that was out of genetics clinics. So we discussed the notion of having small studies that would attempt to optimize the way of collecting this data and evaluate. So for example, a key question is patient-entered versus ancillary health provider-entered types of information. Can we do some comparative analysis between, let's say, information that's obtained from a nurse practitioner or physician's assistant versus what is obtained from a patient? And more broadly speaking, just the setting in which the information is obtained in the context of a clinical visit, what can be ascertained versus what can be obtained, perhaps, outside of the clinical visit? So these would be, I think, relatively simple initiatives to implement and understand better what best practices are. There was a lot of discussion about whether family history tools should be standalone versus ones that are fully integrated into electronic medical records. There are certainly pros and cons to each, and I think we have no idea really which is going to be best for facilitation of integration into the clinical workflow. There was a lot of animated discussion, I think, about using electronic media. I think Mark also mentioned social networking in our discussion yesterday about the possibility of developing a wiki-based type of model that would encourage family members all to contribute to developing a pedigree and a family history. There was also lots of animated discussion about Facebook applications, and I think Jonas overnight has done some additional work on finding out whether that could actually be easily implementable. The notion of an iPad app and also just comparing to the more standard genealogy web pages was we're all elements of potential projects that could be undertaken in this arena. The information interface and education of providers, and I should say patients. I think that the notion that the group seemed to be interested in was not just to deliver information, but also to deliver an educational platform as part of family history gathering. Marin mentioned this this morning, but also what is it that the primary care physicians want versus what specialists want, so I think the high hypothesis was that journalists just want red flags where specialists really want more detailed information about patient history as well as genetics and genomics. So could we just get that information as I think the VA has already begun to do and surveying providers in different settings and understanding what their needs are to optimize their use of family history, and probably a much more challenging aspect would be to create the informatics platform that translates in family history data appropriate for the audience that's seeing the data. So if a PCP environment only wants to see certain things, can you limit what the information content is for them versus for other recipients of the same types of information? And we thought that we could build around the existing tools rather than create something de novo, that we should emphasize the fact that pedigrees should not go away because they really can be informative and educational about the underlying genetics, that the tools should provide the most likely diagnosis, but also to suggest alternative diagnoses that the information should be targeted, brief, and actionable. In terms of the outcomes research agenda, we discussed the desired outcome of adherence to evidence-based guidelines, such as from the USP STF. And we should be able to measure provider behavior in terms of their compliance with these guidelines, and patient behavior in terms of their compliance with physician or provider-based recommendations. There was some discussion about, could this be done in a focused way? In retrospective analyses, knowing the caveats and the biases that might be associated with retrospective studies of looking at how, what happens to patients subsequent to inadequate family history in the chart versus those that don't have adequate family history information in the chart in terms of certain outcomes using the rich data electronic records from places like Marshfield, Geisinger, and Intermountain Health. The idea of doing some prospective studies, a cluster randomization where some practices would practice usual care, others would practice family history-informed care, and measuring, again, the patient-provider outcomes. The notion of specifically collaborating with the VA or the Emerge Network to integrate family history in some areas, but not in others prospectively, and look at compliance with guidelines, behaviors of patients in terms of compliance with diet, exercise recommendations, medication compliance, or achievement of relevant preventer medicine goals around blood pressure control, blood sugar control, cholesterol, and so on. So I think these are all things, many of which can be readily implemented. So the holy grail in some ways is, OK, family history is certainly one type of data, genomic data, molecular information. I think we need to be prepared for a time when all of this data will be resident in some type of data warehouse. And the first thing is to think about initiatives to develop the methodology that assimilate these types of different complex data, all with their different formats and standards so that we can begin to approach what might be the most refined molecular models that can be aggregated to predict disease risk or drug response or whatever the predictive model is aiming to do. We also believe that this requires access to an incorporation of family history with ongoing population study initiatives. And we recommend that structured family history data be incorporated into as many NIH studies as possible, particularly ones that are collecting genotyping or sequencing data, and some of the places where that might happen are listed here. And then finally, we felt that maybe some of the esteemed colleagues in the room or on our working group could actually form some kind of advisory group on family history that could look across the opportunities and make some recommendations where family history information should and could be implemented into ongoing initiatives. And also to probably provide advice to study PIs and steering groups about family history where they may not have that information and be able to inform them about the best ways to collect it and also potentially even to recommend future research in the form of RFAs or the like. So I think that was it. And I hope I did justice to our discussion. If any of my colleagues want to make any comments, please do so. Jeff, a point of clarification. The Mendelian Centers that I mentioned is not my project. We have part of that project at Hopkins, but it's led up by Debbie at University of Washington and investigators at Yale with Rick Lifton as a third site, and we being Baylor Hopkins together. And just so I appreciate your clarification. And I think your press announcement is happening as we speak. Well, congratulations, all of you. I guess I was also curious as to where the OMIM piece is also across the entire, and I meant to spend a lot of time on this, but I think your point yesterday was also to integrate this with the OMIM portion of that. It will be one. So the group will go to and interact with the distributed to the centers based on capacity? So I think the point of our discussion was to potentially collaborate with you to obtain accurate family history data on those Mendelian families that you'll be looking at. So Jeff, yesterday when you gave your talk, to me the exciting parts of that was that primary care folks learned to value family history. And I didn't really see that coming out of the breakout session slides, where to me, I think generating the data that shows value to your target audience is something that's really missing. And if you have to go in and show everybody why it's valuable, then it's never gonna happen. And so another element would be getting some of those talking points with real metrics that could be shared across family medicine. So I think right now people think it's a good idea, but they don't know it's a good idea. And somehow that needs to be conveyed, or maybe it's not as good as we think and we need to just stop pushing it. Right, and I think your point is a good one. We didn't focus a great deal of our time on that point yesterday, but I think it was implicit that we really would like to develop both as part of the implementation science initiatives means to evaluate what provider see and patients see as the value of having this information and as well as also try to come up with some robust outcome measures that could actually be, or outcomes that could be measured in that regard so we can report it out in a way that's meaningful to health systems, to other providers so they really, so it can be translated to other environments where it's not being used. So thank you. I think the other thing that I would add, this is I think in your third project where you're looking at prioritization and how you can be actionable about family history. The thing that I didn't see there was patient control and patient centeredness in the sense that if you have a patient that completes a family history, there may be risk for several things that are identified. So perhaps there's cardiovascular risk, we look at and we say, gosh, we need to talk to this patient about their cardiovascular risk, but the patient may say, I'm really concerned about my family history of colorectal cancer. I think there's an opportunity to use this as a pre-visit negotiation tool about what are we gonna talk about in the 10 minutes we have together and the hypothesis I would have is if we prioritize based on what the patient's concerned about in their family history that we may in fact see a better movement on health behaviors, which some of the health behaviors relating to diet and exercise at lower colorectal risk also happen to impact cardiovascular risk. So I would really like to see a patient centered research focus with these tools. So great point, but not unique to family history as you know, when you see you're a provider and your provider has a list of 10 problems and you have your list of 10 problems and they go like this, you're probably not gonna have an effective interaction. So whether it's family history driven or not, but if your coin is well taken, it should be included. Other questions or comments? Thanks, Jeff. Thank you. Actually if I can ask all of the breakout group presenters to make sure they leave their presentations on the computer up there, it'll help us. Going forward. All right, so next is Alan Howard and Mark, I'm not sure it looks like Alan, pharmacogenetics, genomics.