 Okay, so I had the privilege of chairing the clinical informatics interest group over the last year and a half and I wanted to acknowledge both Casey and Casey's contributions to that group in particular And one of the things is a group that I think the cases already been made fairly well is that clinical informatics is really integral to the success of genomic medicine and so the group has started with a mission of supporting an open forum around exchanging lessons related to clinical informatics and also summarizing and disseminating best practices And so I think you'll see some of the projects that we've been working on are in that realm But we're looking forward to other opportunities and there are many in this area We we have seen essentially summarized in the two previous talks that there are two objectives That are that need clinical informatics to succeed which is having clear actionable representations of genomic risks and clinical guidance within the HR and then also Communicating those results with patients and exchange instruction information between sites of care and one of the things that is not mentioned Which I think is important is enabling long-term management of genomic risks and long-term access to those results So I wanted to reiterate one point which I think the other speakers have made Which is there's been a tremendous run-up in EHR adoption in the last ten years But there's two lines here and the bottom line with this is data reported by ONC is Representing the basic EHR system. So there is truly a diverse Installed base of EHRs in our communities and most of them are basic So the comprehensive ones are represent only about a third of the installed EHR systems and what does that mean? Well, there's actually a list of features which you probably can't read But the one I wanted to highlight was the one that says decision support at the bottom is Lacking in all of the basic EHR systems. So Right off the bat, I think there's a major Deficit in infrastructure to deliver the kinds of guidance we'd like to do within the EHR And we encounter this within our night projects. In fact, one of our sites Has not only had a basic system But they've turned that system over twice in the in a time period where they've been active within ignite So it's the stability of the systems can be an issue as well over time So what we've done as a group in the last couple months is started with this idea. Well, let's let's sort of envision a little more idealized clear compliant genomic data pipeline and This I think echoes some of what Sandy has presented where you have different stages that the data goes through it starts on the laboratory And then we have a stage that we call genomic escrow where It for a large panel test or sequencing result you would Stage the data there and only the data that is actionable would go forward to the EHR but there needs to be a Clear compliance storage over the long term where you could potentially reanalyze that data and produce new findings There are clearly gaps in just about every task within these different stages Probably the only thing that is we do fairly well consistently across sites would be specimen accessioning and tracking But there are lots of opportunities within these other areas and the other thing the other point that I think Sandy made well is that these tasks can be done within the same health system or Many of them can be done within a reference laboratory or even a third party those middle two stages. There's lots of course of Silicon Valley companies that would like to take care of Storing your genetic data for you or doing data interpretation And so the the landscape the commercial landscape is fairly complex in this area Now within ignite we are touching on many of these And with some of the projects that I'll show you Certainly not solving them or trying to boil the ocean, but at least making some progress on some of these areas So as a group we came to came up with these major informatics challenges And I think in terms of the ones that that we've been focused I'm not going to read all of them, but in terms of the ones that we've been focused on lately It's mostly been around delivering clinical guidance based on genetic risks to the right person and place within the workflow But the rest of them are opportunities for us to do to expand our work So one of the major tangible products that the group has helped with is with a site called CDSKB This is a design to be a repository for Implementation related clinical decision support and so it Includes screenshots includes supporting the algorithms documents things that a Site that it was not doing this kind of work could use that to get started now It's not designed at this point to be exportable plug-in decision support That is still technically quite difficult with the range of systems that are out there But this is I think a nice foundation for that kind of work going forward And just to give you an example. This is this the search screen So across all the documents we've collected you can search by Drug name you can search by Jane you can search by disease indexed based on those three categories It were by the type of document that is stored And so within the repository we have Artifacts that are basically screenshots of clinical decision support. This is an example for our own institution showing a Screen that's in the in the middle of the prescribing session where You might want to switch from Prouser girl to from Clopida girl to one of the alternatives based on substitute 19 results And one of the interesting things you can do with this particular resource is compare this to Something that is trying to do the exact same thing at a completely different institution So it's sort of a way to do comparative informatics. This is actually Implemented within an epic system and is much simpler, but is trying to do the Accomplish the same task There's also This clinical decision support that is being generated by ignite sites. This is an example from the guard study related a bell one and And there's a number of other documents and and it's an open site. You're welcome to go browse through them so the other I think nice Activity that SIG has been involved when has been these monthly webinars We've had a number of experts Talk on different clinical informatics topics over the last year and a half Here's some of them And I think our we will continue to do this run this series and so I encourage everyone in the room to participate And we've recently started in the last Couple months looking at this idea of Given the idealized version of the data pipeline. What is actually out at our at our night sites? That's not something we've cataloged yet. And so We are In the process of putting together not only this framework, which is going to look somewhat like the diagram I showed But also a structured data collection to really get at where the gaps are within our own sites Which now cross both, you know, we have academic sites. We have non-academic sites We have community sites and so it's going to be a nice selection And cross-section of what these different clinical care settings look like from informatics perspective The other activity that is related to what we do is we've been active in helping With some of the activities of the pgx group mostly around extracting Drug exposure data or outcome data and doing electronic phenotyping And so this is a project that we've started in the last six months Which is designed essentially to look at How often c-pick level a drugs are being prescribed at a number of different institutions we've collect we've now have a group of 10 or 11 institutions that are participating and so this is going to Allow us to do a couple things It's it's going to look it's going to look not only at the overall opportunity for pharmacogenomics in particular But also look at prescription trends that we should be paying attention to and so this is an example of some preliminary data we put together at our i3p sites and I want to draw your attention to the middle panel where the va made a policy decision to change from Warfarin to one of to Allow their or encourage their prescribers to use the other no acts or novel oral anticoagulants And there's both dramatic drop in warfarin use really in a a couple month time period in early 2015 And so that's the kind of thing that would clearly Impact the opportunity to tailor warfarin therapy Whereas in a place like aurora, which is a large Integrated health network The the warfarin opportunity is still very large where the vast majority of their patients are being initiated on on the traditional medication as opposed to the no acts So briefly summarize we we have a lot of opportunities to study and address gaps in the data pipeline We also have opportunities to support comparative effectiveness activities in the network And one of the Areas which I agree with saying is very important That we could focus on in the future is really more of the user experiencing Of accessing genomic data or interpretations within the EHR Thank you, josh And we will have a discussion moderated by eric borwenko at the university of texas at houston with the discussants terry Klein from stanford university in karin albeck of university of utah