 All right, so next up we have David Sanders. He's originally from Elko, Nevada, went to school in Oregon State and then is now at OHSU and has been joining us. He's been working with Dr. Jacobi as well as Dr. Tabin, Dr. Jeff Tabin. Interesting things about him, despite the fact that you may think his favorite restaurant is KFC, it is not. He's really enjoyed Cafe Rio here and he'll show you what he's enjoyed in Utah as well so I'll let him deliver that message. But the title of his presentation shows electronic health records and ophthalmology and their impact on clinical documentation, so go ahead, David. Thanks, Dr. Swan. And thank you all at the Moran, Dr. Olson, everyone for having me here this month. It's been really fantastic. I appreciate all your teaching and I've had a great time. So since Dr. Swan already started me off with the title, I'm sure a lot of hairs on the back of necks raised when they saw the title with everything that's going on with Epic over the past month and it can be difficult for a lot of people but I bet we can get through it. So I am from Portland. I go to OHSU, a medical student there. And country-wide, Oregon is known, Portland is known for its coffee snobbery, its hoppy micro-brew swelling, and hipster glasses. But there are actually gorgeous places of nature around there. This is a Columbia River Gorge. This is Crown Point, it's about 20 minutes outside of Portland. So pretty places there just like here. Just to give you a little background. So Electronic Health Records, they've been deemed by the Institute of Medicine to be essential for improving the safety, quality, and efficiency of healthcare. I think part of this probably stems from, paper is having a little bit of a hard time keeping up in terms of all the technological advances that are occurring right now, as well as risk of medical errors and the demand for increased efficiency in transmitting information. That being said, ophthalmology documentation is different than other specialties, as you are all aware. There's a real focus on concise paper-based templates and on detailed anatomic drawings and a heavy reliance on actual drawing of findings, which a lot of EHRs don't do a great job with. Which is probably partly to contribute by ophthalmology has really been slower to adopt EHRs than most other medical specialties. It's increasing, as you can see, but still really the minority of practices have EHRs. The federal government is trying to aggressively promote adoption through providing financial incentives and penalties in the form of the HITECH Act. And that's as long as you show meaningful use. So what did we do? This is some research we did earlier this year. This is a comparative case series, a chart review where essentially we had three attending ophthalmologists at OHSU who each had 50 patient charts. This is in three different diseases, AMD, glaucoma, and pigment and carotidolation. And we looked at a total of 300 notes, so 150 paper exam notes and they're corresponding 150 in EHR after the transition. And simply what we did was count. This is our research assistant, Andy. Actually, I have no idea. This child's name, he's a Google image to me. But we just counted. We essentially broke down the exam notes into individual sections, the general exam, the split-map exam, and the fundus exam. And we had signed points for each exam element in the note. So for example, on the split-lamp exam, the cornea exam was deemed to represent two exam elements, one for each eye. Similarly, each of the attending ophthalmologists identified critical findings that were necessary for clinical evaluation of that disease. And we counted them the same way. So I know I bragged about Portland a little bit, but you guys have it pretty good here too. This is the Wasatch Crest this last weekend, biking, incredibly gorgeous. Onto quantitative results. So what we found when we looked at the mean number and percentage of all the exam elements documented. So when we summed up the general exam, the split-lamp exam, and the fundus exam, we found that electronic health records ended up documenting more exam elements overall than paper. And when we looked at the critical findings, the things that were necessary for clinical evaluation as thought by the attending ophthalmologist, AMD was the exception, and this is because there was more comment on sub-retinal fluid in paper, whether it was present or absent, than there was in electronic health records. But glaucoma and pigment and curative lesion still had more documentation in EHR. Also quantitatively, notes were longer overall, and they were for every, I think you suffer, so for every paper note it was only one page, and EHR notes were usually two or more pages, always two or more pages. Qualitatively, just looking at the EHR, as you guys I know have been through this last month, you just see the problem list and medication list, and sometimes it's very challenging to pick apart what is actually going on in the eye. You have to really fare it your way through things sometimes, which is one of the biggest qualitative findings, I think. Oh, sorry. Additionally, what is very, I think, intuitive to most people, and maybe just obvious, is that in paper notes, everyone draws. There are graphical representations of things. There are dots for pigmentary changes in heme, versus in the EHR where there are summary statements and textual descriptions. And this was present across most of the charts we saw, so drawings of the optic nerve heads, but a description here, optic nerves, inferior nacho D in superior erosion OS. And also there were interpretations of findings more often in EHR than there were in paper notes. For example, PRP scars here, we have Xs in the periphery, but here we have overall unchanged good PRP peripherally, so just an example. So a little bit of discussion and take-home points. So overall, EHR notes were longer, and should more complete documentation of exam elements compared to paper. I think knee-jerk reaction from a lot of people is, oh, that's great. That's what we want from an EHR. We want all the information there. At the same time, and so I think a lot of proponents of EHR would say, this would probably lead us to believe that there's improved attention to detail and thoroughness of care with EHRs. On the other hand, I think it's really, Merritt's mentioned that at the same time, there's just this large amount of information that's undigested and hard to access in a clinically relevant manner. So the more information we get, it can make actually those critical findings for treatment and diagnosis harder to find amongst all of the sound. Also, it could be that part of this is a little bit false positive. So a lot of the notes that we saw were brought from like copy-forward, copy-paste, or all-normal, checking all-normal. And while these do save a lot of time, there are definitely drawbacks associated with these documentation strategies. Additionally, the qualitative nature in documentation I think is really interesting. So to give you a little background, we transitioned in 2006 at KCI and here it might be a little bit different, but we didn't have any drawings at all using the EHR and all these 150 charts. And I think that's partly probably because the cumbersome nature of the mouse and drawing retinal findings or any other findings. And I think there's just a learning curve associated with that. So I think I'd like to wrap it up there. Really, I want to, oh, sorry, thank everyone so much. You've been so helpful in teaching and guiding throughout the month. I especially want to thank Dr. Jacoby, Dr. Tabin, Teske, Petty for allowing me to shadow them. Alicia, thank you very much. And all the residents, techs, nurses, and fellows. I also want to thank Dr. Chang and the rest of my research team at OHSU. They've really taught me a lot. So with that, oh, here are my references. I'd like to open it up for questions. You know, there have been times in, I think, especially when they're like fixing the system or doing upgrades to it, where we've needed to go back and use paper charts. But thankfully, since everyone was trained that way, it went fine and they were just scanned in right afterwards. Nothing catastrophic occurred as far as I'm aware. But I'm really low on the chain, as you are aware. Good question. You know, I think those are really good comments and valid comments to bring up. Unfortunately, we didn't look into that in this research, but I think it is a hard thing to get quantitatively since there's so much going on, just like you alluded to. I think it's really important because clearly, that's what we have this for, right? It's for the patients. And there's really little research been done on it at all. I think that is the main, that's like the Holy Grail. Is this actually helping the patients in guiding treatment and diagnosis? And I think anecdotally, I have heard from some attendees, for example, you brought up glaucoma. The glaucoma specialist said, I really like the fact that I can just have all of the pressures, like from the last five years, just lined up and I can look at them all. The visual field, it's not exactly like that yet, but I think hopefully it'll be a little bit easier in the future. I think as we develop the user interface more, for example, I've seen some of the residents upstairs documenting retinal findings using a drawing tool. And I think once these become incorporated and easier to assimilate and less sound, I think hopefully that will help patients. I don't know if that addressed your question completely or not, but I guess in short, I don't know. Go ahead. I am not aware of any studies. I actually don't think there are any. It would be a great thing to study. To give you a little bit, so Portland is starting, they're starting to do that, like Kaiser and Providence are starting to hook up with OHSU on a lot of it, on Epic. And that's starting to happen, but it's not at BC level yet, by any means. But I think that's the direction we're moving in, but it's definitely a slog. Great question. Yes? Yeah, I think a lot of the research has shown that they are able to build more and what you've anecdotally observed is has been observed in literature as well. My main mentor who helped me out with this has done like a larger overarching look at implementation at OHSU and shown that that was the case. Part of the problem though, and one of the things that we talked about on the paper, but I didn't talk about here yet, is that a lot of the structure of electronic health records is built around billing compliance and not around guiding clinical care, which is a huge problem and really needs to be addressed in future, like user interface things and yeah, I tell them. I think that's a great point. Any more questions? Great, thank you all so much for everything. Thank you. Thank you.