This Week Health

Don't forget to subscribe!

Are we serving the internal consumers of Health IT services well? Andrew Rosenberg, the CIO of Michigan Medicine stops by for a two-part conversation on Experience. First the Internal consumer of Health IT services and next, the external consumer of healthcare services. Hope you enjoy.  

Transcript

 This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

Welcome to this Week in Health, it influence where we discuss the influence of technology on health with people who are making it happen. My name is Bill Russell, recovering healthcare, c i o, and creator of this week in Health. It a set of podcasts and videos dedicated to developing the next generation.

Of Health IT leaders. This podcast is brought to you by health lyrics. Have a struggling healthcare project. You need to go. Well, let's talk. Visit health lyrics.com to schedule your free consultation. Today we are joined by someone who I really enjoy having discussions with 'cause he always makes me think and I always learn something every time.

We're together. Dr. Andrew Rosenberg, c i o of Michigan Medicine. Good afternoon, Andrew. Welcome to the show. Thanks. Thanks for having me. This is your first, uh, video podcast, but you were on the show before. I think we met up at the, uh, Becker's conference and did a, an audio only. So people who have heard your voice is the first time they're seeing you for the first time.

This is my first podcast ever actually. Oh, really? First video podcast. Yeah. I, I have, I imagine you're gonna see more and more of 'em. 'cause my, the, I have two production guys and these, the, the millennial production guys are like, uh, you, you need to do video. I'm like, Look at my face. Do you think people want to see my face?

They're like, . It, it doesn't matter. People, people wanna interact with video. So that's, that's why we're doing it. And, uh, so, you know, it's a softball question. Just to get us started, um, you know what, what's, what's one thing your team's working on that you're really excited about it at Michigan Medicine.

You know what, what I find where most of our innovation is occurring is in the backend areas. So we're starting with some robotic process automation around claims denial management, and it's working. So, uh, my hope and expectation is that not only will we show value there where we need it, but again, it becomes the beginning of more of these kinds of technologies elsewhere.

Uh, Similar, similar examples are occurring throughout all of our backend processes, and that's the kind of area that I like to show and showcase to other places. Again, for just that reason, we can be successful there and then really start to take it into the more interesting and maybe more impactful areas of clinical care, research and education.

That's exciting. I, you know, I, I wonder having that resource of the University of Michigan there, do you tap into their, you know, their computer science, their ai, machine learning, all the things they're doing, robotics and other things? Let me tell, yes, we do. And that's one of the great parts about working here.

But let me tell you an item that just came up last week. Here's an example. We, like everyone else, are struggling with how to keep up with the expanding data storage needs. And while the clinical arm tends to be the area where we still have the most of our data storage research is where I'm seeing not just growth, but potentially so much growth that we have no clue how to handle it.

And here's. To me, the best example, and I literally just heard this last week, there is a technology that's actually been around for a while called light sheet microscopy. And what it is is really multiple, very, very high definition cameras, all looking at structures, uh, for example, entire, um, . Fruit fly brain, or now starting to look at entire mouse brain.

But here's the thing that's so fascinating. A single entire mouse, 12 camera, 10 or 12 camera light sheet microscopy could use up to six petabytes of data. Single study.

We are going to have to do something new in compression, in data handling and data management because if we start doing that, even at the level of invertebrates and mice, when we start trying to do that for humans, we don't even have a number yet of what the data needs are gonna be. Those are examples.

High def, uh, video studies for two weeks, and of course, precision medicine. We're just completely blowing open the need around data storage, and it's not just a money issue anymore. It really is gonna be fundamental changes in technology to do these kinds of new things. And that's an example of something happening in Michigan Medicine, but also on campus that we're gonna have to come together with engineers, computer science, school of information, and frankly, even new areas because it's a complex problem and it needs diverse people to fix it.

Wow. Yeah. Petabytes per image. I, I can't even imagine. I mean, and it's, it's per study, just I'm clear. But, but if you, if you, if somebody's gonna say, well, you know, just move it to the cloud. I, I'm just, I'm doing the math of moving that to the cloud. That's, that's really, that's a lot of money. And where we've had these earlier examples of this, it was the place where compute and storage have to be close because of all the IO that goes on.

And at those scales, it's, again, it's just an example where in cellular developmental biology, molecular biology, these are new examples where . Uh, we have faculty now doing that on premise. So it's not just, it's coming in a couple years. It is here and now how do we deal with that? And of course, I think that is translatable back to the clinical area when we start doing high def, long-term video.

I. For patients, families, things like this, we're, we're going to have to be tackling some of these problems. And they're, they're, they're wonderful problems because they're gonna require a lot of different people to work on them. Yep. Well, you know, the two main topics of our discussion today are gonna be, um, well, it's one topic, really, two directions.

So the, the topic is, is experience. So we hear it over and over again. I was gonna talk to you about disruption, but I didn't know where that was gonna go, , because, uh, I know that you. I, I was with some CIOs and you were one of 'em when we were talking about disruption and just a bunch of us was like, if we hear the word disruption one more time, it's, that's too many.

Transformation is the new word. Transformation. Uh, so we're gonna talk about experience, which is another one of those words which we're hearing over and over again, but I'm gonna head in two directions. One is the, uh, clinician or physician, um, physician or clinician experience, and then the second being the consumer.

So let's start with the internal customer. Experience and, uh, who, who we are all very aware has, uh, had a very difficult time through meaningful use of, um, all the things that we've sort of thrown at them. So let's start by setting the baseline for the discussion. Um, uh, how are we doing today? How are we doing today?

Are we doing better with the clinicians, administrators, physicians, or are we just sorta um, just sort of getting by? You know, this is, this is a moving target and my answer that I've said at conferences, what you've heard me say is, I think we're doing a lot better. The problem is that because the target's moving so quickly, your point of reference will, will bias your, um, your answer.

So, The story that I like to give around this is that when I started as faculty and I was doing research in the I C U, that really required technology. It required attachment to network, it required decision support tools. It was a problem for me. I've used even more colorful language around it. And the, the problem was the people working in it were superb people.

They were just air gapped. There were no translators. There was no clinical scientist type people who bridge the gap of both classic it, classic information services, classic technology, and the various clinical needs. And so I think certainly in 1998, If you ask clinicians, researchers, educators, how is it supporting your needs?

I'd say it was terrible. They were building backend, uh, systems. They might do some very specific custom software for an individual that was wonderful, or sometimes in very rare places, you know, the, the, the well-known informatics units within the, the country, you actually had some pretty good enterprise level work, but it was still siloed.

Now, You couldn't find a physician, a nurse, a pharmacist, a respiratory therapist. Who doesn't on a daily, if not hourly point, see the connections with it, with devices, whether they're general compute, whether they're communication devices, whether they're smart boards, uh, digital signage, you know, uh, let alone some of the more advanced things like analytics, uh, sensors and things like that.

The problem is, because we're getting used to those so quickly, All the problems associated with it. Take the E M R. You know, I was, I was at a conference a couple years ago and people were talking about all the problems of interoperability, and I pointed out with HIEs that exist now, we have much more interoperability, much more data sharing than we ever did in the paper era.

The problem is it's not nearly meeting the demands that we now have. So I think this conversation has to be one from perspective, uh, also one from the relative time that we're talking about, because I think it's such a moving target, probably we will never be satisfied in answering that question. We always want the next thing.

But we, so, you know, some people have made the case, Hey, let's just go back to paper. But, but we needed to digitize, I mean, this, we couldn't be the only industry that didn't digitize. Uh, what will some of the long-term benefits be to having digitized healthcare? Uh, We're, we're chasing. Um, you know, I mean, I, I spend a lot of time on social media only because, you know, we promote the show through social media and I engage, uh, listeners through social media and I see these things.

You know, somebody will post this great story about some success story, and then you'll have a couple people come down below and say, you know, there's no return for all the money we spent on. On digitizing healthcare and there's no benefit and, you know, people aren't any healthier and we're spending more money, um, you know, do you think we need it to, to digitize healthcare?

And what do you think the benefits are gonna be? It's, it's such a massive question. I would argue, uh, pick an area of real interest and start looking at where, ask the question. Would people truly go back to paper? Um, patient portals. For the most part are,

are it, are, are liked to loved by patients, they want more. Um, but the ability to access labs, the ability to, um, reorder medications, I would say is in and of itself an immediate . Positive experience for most patients. At the same time, that patient portal is added to an in baskets deluge. That's not dissimilar from, uh, our email and the workflows to handle it on the other end are enormous.

So there's an example of a digitization that I don't know that many patients that would prefer to go back to . Having to call and wait for written prescriptions are going to pick them up. And I would also argue that even among most of our recalcitrant physicians, the ability to see patient data in the h i e, for us, it happens to be epic.

That works very well, has been. Almost universally positive. Whereas again, the fact that they can only get part of that data and not all the data is seen as either a negative or a minus one. I would argue it's still just an incomplete aspect. Um, think about some of the biggest changes that are going on in chemotherapy immunotherapy right now.

All of those are based on some degree of digitization of data. This happens to be biologic data, genomic data, marker data. Um, Uh, when we think about imaging, uh, some of the emerging, uh, AI for, um, ruling out, uh, common but still potentially deadly. Things such as pneumothoraces on chest x-rays, let alone where it will be going.

The digitization of, uh, pathology, the ability for. Radiologists to increasingly work from almost anywhere in the world doing higher value reads at night instead of waiting longer time. I mean, we can just continue with these examples and every one of these examples, which ultimately are digitization of either the textual data, the verbal data, the imaging data, the molecular biology data, all of them had advances Now that I can't imagine anyone.

Remotely going backward to paper or analog. Again, the problem tends to be, well, if we can do some very basic AI for chest x-rays, why can't we do it for high res ct? Well, because there's steps to get there. Why can't I get all of my images right away in real time? Well, then we get into problems of storage and retrieval.

So, I see these examples everywhere and almost always I see a duality of, you probably wouldn't go back to paper. I, as a clinician, wouldn't go back to paper in the I C U or the OR ever. But I can also list all sorts of things and now I'd like to see done even better. Right. And you know, and we hear these success stories on the clinician side.

So on the internal, internal consumer side, we hear these stories all the time of physicians who are much more productive. They can do, um, you know, they can do reads from anywhere they can. Uh, you know, there was a, uh, Uh, Dr. Wachner shared a story of a U C S F physician who was able to, to essentially improve his productivity almost, almost a hundred times, and just the ability to, um, to just handle things because they, they built some custom screens within the Epic environment.

They were able to bring the data in and he, because, because it was customized to his workflow, he was. Uh, uh, and he just told the story of how much more effective he was, uh, across the board. Um, but the trick seems to be, you know, those, those success stories are not as common as I think we, we would all hope, I mean, you could find a couple in each health system, but what, you know, what's keeping us from, uh, you know, from, from proliferating those stories or having them impact more of the health system.

Uh, I, I, I think it's a political question. It's an emotional question. It's a age question. Um, you know, all of my training, all of my training, my medicine, my critical care, my anesthesiology training was essentially done in paper. Actually, my, the last part of my anesthesiology, we built our own anesthesia information system here at the University of Michigan, and so that became digital.

Pretty early. Um . If you've trained, if you have all that muscle memory in paper and analog work, I think it's very hard to expect anyone to just fully embrace a digital workflow and in a digital environment, let alone, uh, how you engage with it outside of your profession. Um, so I think when we hear about stories of either productivity gain or productivity loss, I, I, I didn't say this when we were implementing our E M R, but I certainly felt it, and I, I, I said it to a few people.

I. In many ways, we were building these for two generations from those that implemented. And of course, for anyone who has spent all the time and effort to become a physician or a nurse, or a clinician, or an educator or whomever, and then to hear we're spending, like you said earlier, all this money and not getting an immediate return on investment.

I take a much longer view in general, maybe somewhat philosophically, but very pragmatically. Then we also don't train nearly enough to take advantage of even the basic technologies we have now. Most of us who are practicing right now, spent about 10 or maybe 12 hours training to use the E M R. And very few people have gone back to take another few hours or a half a day or even a day.

Those places that are doing that universally, the people coming outta that training are saying that was extremely worth it and increasingly demonstrating some outcome improvements. We're doing that with our Home for Dinner program at U of M. It's based on Kaiser Permanente, it's based on Stanford's, it's based on these other examples, but.

We're getting at least 50% reductions of excess at night work because we're just training better. So it's a complex question with a lot of complex answers, but I see efficiencies gained when people embrace, when people have a number of their colleagues who are working both on the clinical side and on the design and implementation side.

And as the technology continues to improve, like voice, voice recognition, we'll see more of these productivity gains and we're already in some, uh, focused areas. So tell me a little bit about, so you're the second or third person who's talked about training. And, uh, it, not that it's a magic bullet, but that training has been, uh, a significant, uh, training and the physician builder have, have been two things that have really contributed to, uh, uh, satisfaction, uh, around the E H R effectiveness, around the E H R.

Talk to us about your Home for dinner program and, and why, why do you think that works and, and how is it working? So our Home for Dinner program, which the title I believe is linked from that, that Stanford coin, thank you Stanford. Um, and a lot of the elements are what we learned from Kaiser that learned it.

Two years after their implementations. Thank you. Kaiser. Uh, there's only common sense peers and colleagues train each other in the more efficient use of the tools. These are. Colleagues who have spent the time and effort have the inclination to know those tasks, that waste time, that are amenable to efficiency and train each other how to do that.

Uh, preference lists, uh, modifications of the, um, of the screens that are most commonly used to provide some degree of automation. Those things that are most commonly done and often. Like so many things in life, it's these small incremental improvements that add up over time a day. You know, saving a half an hour a day might not seem like a lot until you realize that over a month and over a year you've seen you've saved several days worth of nighttime work, things like that.

Um, I would still say this is education 1 0 1. Most of these training that a clinician to clinician, a nurse to nurse, a pharmacist, a pharmacist, a respiratory therapy to respiratory therapy, uh, respiratory therapist will to each other. These are things where I know what your issues are, and I know if you did these two or three things, you would shave this kind of time.

Thank you. You know, so there's peer trusting, there's very simple things. And then once you get that done, then there's more advanced. Sometimes it's the teamwork, uh, inbasket, I talked about that earlier. Everyone who deals with in baskets know that it's really a team effort. Um, it's the people who can be first blind.

To be not just accepting the messages, but not just forwarding it on because that didn't save anyone any time. It's empowering people to work at the top of their license. It's empowering people to do new work because the technology allows them to in a different place. So there's some really basic common sense things going on.

How much, um, you know, how much customization do you end up doing to the E M R for specific? Uh, specialties around those things because you, you don't wanna, you don't wanna mess around with the foundation or whatever the, the core system is, depending on the e h r provider, whatever they call it, because then your upgrades become challenging.

Um, but it's also been sort of shown that if you can create, uh, more custom, uh, screens and more custom, uh, information to that physician at the point of care that they're . They're able, effective, I mean, how, how do, how do you sort of view that and how do you address that? Well, this is one that everyone has implemented goes through and let me give you both a clinical, but then lemme give you a nonclinical, more administrative IT example that I think might be even more interesting.

So the clinical one was, You know, for these big sociotechnical projects, like all the E M R implementations or V n A and PAX replacements are, there's a certain, um, professional and then there's a certain emotional and political side to it. The. Pure technology answer would've been to do out of the box as much as possible for one of the reasons you brought up the ability to upgrade more efficiently.

That's usually a loser right from the start, especially with physicians and unfortunately a little less so for nurses, but they have just as much inefficiency and, and the reason I say that is that. Most places I've talked to and been at the customization tends to be more for the physicians. When I would argue it, it could as much be as efficient or even more so than for the nurses and some of the other care team, but they tend not to get that kind of attention.

So the way we did it to try to um, uh, find as much of a moderate middle ground was the following. We have about 88 services in Michigan medicine, in the health system. Um, uh, uh, Uh, uh, pulmonology, uh, heart failure, cardiology, uh, trauma surgery, general surgery, things like that. For each one of those services.

We had a physician or provider champion, and they were allowed a certain number of order sets for that service. I forget exactly how many we, we allowed. It was either five or 10. It's been now. Nine years, so I'm forgetting. But so we allowed a certain degree of customization because adult, uh, pulmonary is actually very different than pediatric pulmonary.

Same thing for gastroenterology. Uh, hepatology is different than general gastroenterology. Adult urology is an extremely different thing than um, uh, uh, uh, G y n urology. So based on the services, we allowed some degree of customization, but then everyone in that service had to use. Those, um, approved order sets.

So it was an attempt to both customize but limited to a service. So I've heard some places where individual physicians could have their order set. We, we never allowed that from the get go, and so that allowed. A reasonable amount of customization to preserve workflow, to preserve those things that were unique to the University of Michigan as opposed to Mayo or Harvard or Stanford or whomever.

Um, and it also then allowed upgrades to not have the burden. And I would say we did it about as well as one could do it. Now, let me take another example that is a little bit more modern. Uh, we've implemented ServiceNow for one of our big I T S M, our IT service management tools. We have. Essentially taken that vanilla out of the box.

Again, politically, I could help manage that because it was all in my IT group and I could, I could manage that challenge to individuals who say, well, I really need to be able to do this based on how we used to do it in our older system and say, we're going to go as much foundation as possible, and that has allowed us to succeed in some areas where other places that did

Extreme customization really have challenges. Where that's gonna play out for us in a new way will be when we migrate from our E R P system in the next, you know, several years. We start that project into a cloud-based e r p. That will be a very interesting challenge again for try, try to take as much foundation as possible.

I bring these examples up because we talk a lot about the E M R. But the reality is we have hundreds of applications that our providers use, and so we have the same issue You brought up, how much customization versus how much vanilla we take. And then when you start to add SaaS onto it and the frequency of upgrades that we're gonna have to take, I think we're all going to find we will migrate more and more away from customization and probably migrate to more

Foundation, however you wanna define that with an interesting need to educate and to frequently reeducate people on upgrades and new features as opposed to rely on the age old work of customization. I. That concludes part one of our discussion with Andrew Rosenberg, c i o from Michigan Medicine. I wanna thank Andrew for coming on the show.

It was such a great conversation. I appreciate his thinking and his approach to health it. I also really appreciate, uh, the time which he gave us on the show. You'll have to come back next week to hear the second half of the conversation where we turn our attention to the external consumer. We spent this week talking about the internal consumer of health IT services.

Next time, uh, we get together, we're gonna, uh, uh, share with you the external, uh, consumer of healthcare and health IT services. This show is a production of health this week in Health It. For more great content, you can check out the website at this week in health it.com or the YouTube channel at this week in health it.com/video.

Thanks for listening. That's all for now.

Thank You to Our Show Sponsors

Our Shows

Newsday - This Week Health
Keynote - This Week Health2 Minute Drill Drex DeFord This Week Health
Solution Showcase This Week HealthToday in Health IT - This Week Health

Related Content

1 2 3 247
Transform Healthcare - One Connection at a Time

© Copyright 2023 Health Lyrics All rights reserved