[MUSIC] Hi, I'm Bruce Darrow. >> I'm Nicholas Genes. >> And we're going to be talking today a little bit about our vision of what medical and health technology is going to look like in the future. The EMR of tomorrow, if you will. So, what do you think the EMR of tomorrow will look like? >> Well, it's exciting to think about and it certainly as we've seen like a revolution in consumer health technology apps that track your biometrics and then so forth. It's become a little bit of a hobby to look at how electronic health records, tools that doctors and clinicians use, how they are evolving. The pace is not the same unfortunately and I've been waiting for a lot of innovation in this space but it is a slow going process. It's just been an effort to get the country wired and using digital health tools, and then just getting them interoperable is another massive effort. But it is exciting to kind of think about where the field is headed and what kind of tools we'll be seeing in the future. Having said all that, I am looking forward to increased intelligence, more context aware, alerts, and just even the display of information. I think about Google. When you enter a Google search term, you end up with a list of links, but that list is somewhat specialized for you. It's based on your location. It's sometimes based on your past searching history. Google tends to know a lot about your preferences. And so, you get a list that Google thinks is the most appropriate for you at that given moment at that given time. And when I work in the emergency department, I imagine that in the future our electronic health records will be presenting information about patients also in a kind of context aware manner. Where a patient's medication list, a patient's past history will be prioritized based on their chief complaint or based on even the time of day or based on recent test results. For example, if we're all going to have pretty voluminous, thick charts in the future because all our records are going to be digital and we're going to have a lot of past encounters. So if a patient is coming into the emergency department with an ankle sprain, maybe the electronic health record would de-emphasize information from years ago about a pregnancy. Or years ago about an irrelevant procedure or surgery and instead would prioritized similar kinds of visits and medications that were used at that time. That's one example of maybe more context aware visual displays. >> And in terms of the information being displayed, I think that we have the sense that we're only starting to surface of what can be done with decision support. So, is one thing to be able to say hey, you're prescribing aspirin to this person, maybe you should be aware of the fact that this patient is allergic to aspirin. >> Mm-hm. >> That's very basic decision support or you were giving a medication that reacts with another medication that this patient is taking. But, we theoretically should have the ability to say, hey, you are prescribing a cancer medication to a patient who only has a 30% likelihood of responding to this regimen. And, there are better options out there. Or maybe the x-ray that you ordered for this patient is not the right x-ray for this kind of condition and you're better off with a CT scan. >> Yeah. So often I feel like I'm trying to do battle with the electronic health record. The pop ups that are coming up are based on key words, like you said. Like, you've entered this, therefore you must see this warning. And instead, it's not a context specific kind of a guide that can help you along and help you make the right decision without disrupting your workflow too much. I actually, I recall with some trepidation Microsoft Clippy, back in the day. This was- >> That was the little paper clip assistant that everybody hated? >> Yeah, everyone hated Clippy, and I did too. I don't want to say Clippy was the paragon of excellence. But in Microsoft Office, Clippy was monitoring your activity and trying to detect it looks like you're writing a letter, here are some tools about writing letters. It looks like you're making envelops for addresses, and so forth, and it would try to point out aspects of Microsoft Office that you might not know about. We don't need that kind of very basic support, but I give Microsoft some credit for having a very context-aware monitor that was kind of looking at what you were doing and trying to give suggestions. When I look ahead to actually Regenstrief, one of the cutting edge informatics groups, they have a next generation version of their electronic health system that has something a little bit like Clippy. It's just this signbar along the side of the screen and it's constantly seeing what you're up to, whether you're charting, whether you're placing orders, whether you're browsing old orders. And it gives these very non-obtrusive little suggestions. If it looks like you're placing orders to look for a pulmonary embolism, it'll display some calculators to help you estimate the likelihood of pulmonary embolism. And it'll have even links to articles that support your decision making. And if you choose to ignore it, it's off there to the side. It's not obtrusive. It's not disrupting your work flow. It's a kind of context aware guidance that I think a lot of clinicians need, or at least would appreciate. >> One of the things that I think we'll also be seeing is an evolution away from sort of the old way of doing things towards a new way of doing things. And at this point in time, we're barely ten years into the smartphone era. >> Yep. >> And the smartphone was able to occupy a space that could be completely re-imagined. >> Mm-hm. >> Nobody walked around with a desktop phone. So people didn't have to have this idea that your iPhone was going to function like a desktop phone. And they could start looking at it as a way of saying not only were we bringing in communication, but we're bringing in information and it can connect to this internet that already exists. We don't have to rebuild that. But I think that a lot of what the current and previous generations of electronic health records have done is say, we are in the space that doctors and clinicians have been working for decades, if not hundreds of years. You go to a doctor and they sit, he or she sits at a desk. And, she writes down things and she dispenses medications and knowledge. And, because the electronic medical records is in that space, it's sort of been constrained by the experience that led up to it. And a lot of it was built on the preferences of doctors who said, if I wrote it this way before, I want to do it that way on a computer, and the way I think about medications has to be the same way, this way. And I think we're now getting to the point where we realize that when the information is electronic, when the information is accessible, distributable. Then you can bring patients into that conversation in a way that never could have been done in the 1970s, 1980s era of medicine. And I think that gives us an opportunity to think over the next ten years of, am I building my EMR in a way that make sense not just to the clinicians but for patients, as well. >> That is really intriguing. And I'm, thank you for sharing that. I'm tempted to try to imagine what that would look like. Maybe a more collaborative environment or opportunities for patients to contribute data. >> Well, a combination of things. So first of all, there are already initiatives like OpenNotes where a physician who is writing a note when a patient sees them in the office can share that information with the patient. And the patient can look at it the same way that he or she would look at bank statement. They go into the computer and they don't have to wonder, did the doctor give me this piece of advice or that piece of advice. Is the doctor saying and thinking things about me that he or she is unwilling to tell me. It opens up that communication, holds the doctors accountable to the patients. It makes the patients more informed, and it gives the patients an invitation to as you say, contribute the information. Say well, actually, it wasn't my left knee. It was my right knee that I injured playing football. And you may think that I'm still taking that medication, but I actually stopped that medication on the advice of another doctor three weeks ago. So once the information is there, that never would've occurred in the doctor's office in the 1980s, but now it can if we allow ourselves to go down that road. >> Yeah, one of the opportunities we're looking at in the emergency department is patient reported pain scores. We collect pain data just like we collect vital signs. But, obviously, patients are very interested in having up-to-date pain information, because if they're in a lot of discomfort, they want to be able to report that and get appropriate therapy. And yet, there's a lot of hurdles to just building this kind of conduit so that patients can enter data on their phone or tablet and have it displayed in front of a nurse or doctor so that we can make appropriate decisions. But yeah, maybe in the future, that kind of interoperability and sharing will be possible. >> One of the things that you mentioned in passing I think was shared decision making. >> Mm-hm. >> And, this is an area of opportunity where health information technology platforms can potentially support this. The idea behind shared decision making is you're moving away from a model where the doctor says to a patient, you should do this, you should get this test, you should get this surgery, you should take this medication. And the patient says, yes sir, yes ma'am. Instead it's more of a discussion to say, well for example, a patient will come to me as a cardiologist and say, I have chest pain. And I can say, based on your description, your risk factors, your history, here's what we know about the likelihood that this chest pain is due to a serious heart condition. Here are our options, we can do nothing, we can try medication, we can try various forms of testing. Here are the risks of each of those, here are the benefits, and we can have a discussion of which one is most consistent not only with what's medically beneficial to you, but your own sense of risk and knowledge. >> Yeah, yeah. This reminds me of a demo of Watson and when we talk about the future of electronic health records- >> That's IBM's cognitive platform, Watson. >> Yeah, yeah, Watson. And Watson plays Jeopardy! And Watson can do a lot of things, but IBM is really focused on health care for Watson, for applications. And if you just look at their commercials, it's seems that they're using almost like the old Greek oracle approach, where you feed information to Watson and you get an answer back. But, I think they're doing a disservice, Watson is actually a fair a bit more intelligent than that and can actually monitor conversations. And actually, not just suggest therapies or suggest tests, but even suggest questions to ask to elicit a better outcome for the patient. It's a little bit of that shared decision making but the demonstration I saw was with a cancer patient where the clinician was going over details of the patient's biomarkers and which drug would work for which kinds of cancer, etc. And at one point, Watson just chimed in and says, is the patient a musician? And I remember thinking, where is that coming from? That's out of nowhere, but one of the chemotherapy agents that was being discussed was actually pretty effective but also known to cause hearing loss and known to cause a lot of damage to the ear. So in a shared decision making model, you can imagine some patients would rather go with maybe a potentially less effective cancer regimen if it would preserve their hearing. And that's the kind of the thing that maybe an intelligent assisistent would be able to chime in and help with. >> And as doctors, I think we are frequently guilty of assuming we know what the patient is going to opt for. That they're going to want the best cancer therapy available even if it has a high likelihood of deafness. But, we don't know until we ask them. >> Yeah, yeah, true. And so, collecting that information from a patient and having it available at the time of decision making is a good role for electronic health records in the future. >> So in terms of the way that technology is moving, we're in the early days now of the smart home. >> Mm-hm. >> Where you can have a home or an apartment that is controlled to have central control of thermostat, and did somebody break in, and I can announce to the walls I want pizza. >> [LAUGH] >> And do you think that with technology and health IT in particular, we are moving into the era of the smart patient? >> Well, certainly I feel, maybe in New York where we're a little bit, I like to feel, on the cusp of things. I do feel a lot of our patients are very informed and very tech savvy and collecting this kind of data already about themselves. And they want help making decisions about that. And sometimes I feel medicine hasn't caught up to where the patient is, at least these smart patients. >> And I think that right now, we're still on a sort of a fact finding period where the people who as patients undergo monitoring and do so to a large base on faith. >> Mm-Hm. >> And they think that its information that will make them healthier and smarter and better in the long term. And I think that's the promise and the hope, but we'll see if that bears out over the course of the next couple of years. But I think that health IT will be more incorporating of over time the ability to take care of the patient when the patient isn't actually there. >> Yeah, remote diagnostics and remote therapy. Treating the patient where they are and where they would like to be, so that they're not forced to make these treks to the office. But yeah, and I'm hoping that with increased interoperability and this secondary use of data, that it becomes easier to share these step counts and heart rates. And be able to unmask potential new conditions and new disease states, or sub-clinical states so that we can detect disease before they become really bothersome to the patient. Right now, if a patient were to tell me that their steps count were declining or that their heart rate was showing increased variability, I wouldn't know what to do with that information. It might be a red flag. It might prompt more vigilance. But, I don't know how I would intervene to turn that around just yet. And hopefully, with aggregating this data and putting good data scientists to work on figuring out what it means, we'll have some actionable information. >> Right, doctor's haven't been trained to deal with information that they never had before. >> Yeah, yeah, right [LAUGH]. We like to put things into the categories we're comfortable with, familiar with. And that's why EHRs may look the way they do [LAUGH]. Well, hopefully there's a lot of promise and sometimes the degree of progress is not what we're hoping for. But when I look ahead to what electronic health records can be in 5, 10, 20 years, I'm pretty optimistic and hoping that we get there soon. >> Yeah, I saw a quote that we always overestimate how far we're going to get in the next two years but we underestimate how far we're going to get in the next ten. >> Yes. Bill Gates said that [INAUDIBLE] [LAUGH] >> I think that would be an exciting ten years. >> Thank you very much, Bruce. >> Thank you. [MUSIC]
