Top of Mind for Healthcare CIOs #4 – Automation in All Areas of Healthcare

Speaker 00:00:03 Today in health. Speaker 00:00:03 It, the story is top of mind things for healthcare CEOs. Speaker 00:00:08 And this is a top five list based on my recent interviews today. Speaker 00:00:12 Number four automation. Speaker 00:00:14 My name is bill Russell. Speaker 00:00:15 I'm a former CIO for a 16 hospital system. Speaker 00:00:18 And creator of this week in health, it a channel dedicated to keeping Speaker 00:00:21 health it staff current and engaged. Speaker 00:00:24 If you were listening to the series, you know, we have four new shows, four shows, Speaker 00:00:30 or I'm sorry, four channels for next year. Speaker 00:00:33 If you haven't already hit this week, health.com/shows. Speaker 00:00:37 And subscribed to all four of the new channels. Speaker 00:00:41 We would love for you to do that. Speaker 00:00:42 If you haven't been listening, go back to Monday, catch up on the Speaker 00:00:45 five top of mind issues for CEOs. Speaker 00:00:48 And let me tell you quickly about the four channels that we have this week. Speaker 00:00:52 Health news. Speaker 00:00:53 His way to stay current this week health conference, we are going to have keynote Speaker 00:00:57 interviews and emerging products. Speaker 00:00:59 This week health community is where we hear from you about interesting Speaker 00:01:03 solutions to problems facing healthcare. Speaker 00:01:06 From people who are solving them. Speaker 00:01:07 For example, I have guests hosts for next year. Speaker 00:01:12 And we have people like Craig, Richard Advil, and a angelic Russell data Speaker 00:01:17 scientist who are going to host their own shows with their own networks. Speaker 00:01:20 10 to 20 minute episodes. Speaker 00:01:22 So I'm looking forward to that. Speaker 00:01:24 That's a new new venture for us. Speaker 00:01:26 And we also have this week health. Speaker 00:01:29 Academy. Speaker 00:01:30 And this is where you go or send people to learn about the intersection Speaker 00:01:34 of technology and healthcare. Speaker 00:01:36 You can sign up again at this week. Speaker 00:01:37 health.com/shows. Speaker 00:01:39 All right. Speaker 00:01:39 Top five. Speaker 00:01:40 Top of mind issues for CIO is based on the interviews that I did at the health Speaker 00:01:46 conference, the chime fall forum and at the healthcare to healthcare event. Speaker 00:01:49 And. Speaker 00:01:50 Just phone conversations and the various zoom calls that I do Speaker 00:01:54 with CEO's throughout the week. Speaker 00:01:56 Number one was labor staffing. Speaker 00:01:58 Number one, issue going into 2022, number two, cybersecurity, number Speaker 00:02:02 three, all things, digital, digital innovation, and all the cool things Speaker 00:02:06 they're doing in digital today. Speaker 00:02:08 We're going to talk about automation. Speaker 00:02:10 And tomorrow, Friday, we're going to talk about care venues. Speaker 00:02:12 So let's start automation. Speaker 00:02:14 Let's start with what is driving automation? Speaker 00:02:17 Number one is that competition for labor in it? Speaker 00:02:22 All right. Speaker 00:02:22 So we have an it labor challenge right now. Speaker 00:02:25 Number two, the need to move up the stack. Speaker 00:02:28 We want our staff to be working on higher level problems, and Speaker 00:02:31 we're trying to offload some of those mundane, repetitive tasks. Speaker 00:02:35 And I'm still talking about it in this case, but that could also be Speaker 00:02:38 applied to the clinical setting. Speaker 00:02:40 Number three thing, that's driving automation, labor, shortages, and Speaker 00:02:44 burnout in the clinical world. Speaker 00:02:45 So we're looking at clinical automation as well. Speaker 00:02:48 A number for the advancement of technology, right? Speaker 00:02:52 Computer, vision, machine learning, AI is driving it. Speaker 00:02:55 And if I had to think about it, Speaker 00:02:56 And this isn't really on my list, but was probably a financial pressure. Speaker 00:03:00 That's pushing automation as well. Speaker 00:03:02 More with less. Speaker 00:03:04 All right. Speaker 00:03:04 So let's take a look at three areas that are seeing automation. Speaker 00:03:09 Number one, it automation to administrative automation Speaker 00:03:13 within healthcare and number three, clinical automation. Speaker 00:03:16 All right. Speaker 00:03:17 Let's start with the area. Speaker 00:03:18 We're all familiar with it. Speaker 00:03:19 Automation. Speaker 00:03:20 This is a new it's been around for quite some time. Speaker 00:03:22 Most are using VMware and Microsoft in this capacity. Speaker 00:03:26 And it's had some automation capabilities built into them for awhile. Speaker 00:03:30 You know, but things like a puppet get lab red hat and chef Speaker 00:03:35 are also prevalent in this space. Speaker 00:03:37 We've done it on site. Speaker 00:03:39 But where we see this more often as in the cloud and the cloud infrastructure plays Speaker 00:03:43 that are out there, it's built in, spin up a cluster for load balancing, click done. Speaker 00:03:48 So it's pretty cool that way. Speaker 00:03:49 And now we're applying it to our in-house. Speaker 00:03:52 Data centers as well. Speaker 00:03:53 We're even doing it in a multi-cloud environment and doing it across clouds. Speaker 00:03:56 Very very interesting concepts. Speaker 00:03:59 Let's see. Speaker 00:03:59 Here are the concepts. Speaker 00:04:00 The concepts are like infrastructures code, task, orchestration, Speaker 00:04:04 and event driven workflow. Speaker 00:04:07 These are things that are entering the vernacular. Speaker 00:04:10 If you will. Speaker 00:04:11 If it hasn't entered yours, here's a quick primer on them. Speaker 00:04:15 So infrastructure is code. Speaker 00:04:17 I just, I discussed this on the show with with several CTOs over the past year. Speaker 00:04:23 Imagine knowing that in the event of a ransomware event, You could execute code Speaker 00:04:28 that can rebuild the infrastructure in minutes instead of days or even weeks. Speaker 00:04:33 Right? Speaker 00:04:34 The code holds the key to rebuilding servers and the proper Speaker 00:04:37 configurations and the software. Speaker 00:04:39 Yes, you have to manage it like code with release schedules and Speaker 00:04:42 repositories, but in the end, this is an amazingly powerful tool. Speaker 00:04:47 And we talked to one CTO that talked about the fact that they could spin up Speaker 00:04:50 their entire epic environment in minutes. Speaker 00:04:53 With a click of a button. Speaker 00:04:55 And that again. Speaker 00:04:56 An incredibly powerful tool, especially in the Dr. Speaker 00:04:59 The disaster recovery business continuity world. Speaker 00:05:02 Task orchestration is just what it sounds like within many it Speaker 00:05:05 organizations, there are hundreds of repetitive tasks being done every day. Speaker 00:05:10 With the crunch on labor, imagine being able to automate a bunch Speaker 00:05:16 of someone or even a team's job. Speaker 00:05:18 So that you can free them up for some of the work that's going on done. Speaker 00:05:22 Plus the added benefit that we have to talk about is giving Speaker 00:05:26 them more rewarding work. Speaker 00:05:29 It's a significant retention when. Speaker 00:05:31 And then finally event driven workflow utilizes things like API APIs. Speaker 00:05:36 That are built into much of modern infrastructure to create sophisticated Speaker 00:05:40 workflows based on triggers. Speaker 00:05:41 Let's see. Speaker 00:05:42 I imagine if your container, I love this actually, now that I Speaker 00:05:46 think about this, imagine if your container could detect an attack. Speaker 00:05:50 And trigger a process that destroyed the container and rebuilt it and move Speaker 00:05:54 the workloads over to the new container. Speaker 00:05:56 Leaving the attacker, always having to adjust and refactor. Speaker 00:06:00 You know what they're going to do. Speaker 00:06:02 They're going to go somewhere else. Speaker 00:06:03 They're going to go somewhere else. Speaker 00:06:04 That's a lot easier. Speaker 00:06:05 It's pretty cool. Speaker 00:06:05 Right? Speaker 00:06:06 So event driven workflow. Speaker 00:06:07 So there's where those were the three. Speaker 00:06:09 Primary things I would think about in this area. Speaker 00:06:11 Infrastructure is code TAC. Speaker 00:06:13 Task orchestration and event driven workflow. Speaker 00:06:16 Like I said, this is just a primer. Speaker 00:06:17 Not going into big detail. Speaker 00:06:20 All right. Speaker 00:06:20 Let's talk about administrative automation. Speaker 00:06:24 Robotic process automation is the word. Speaker 00:06:27 It's the terminology that gets used in this area. Speaker 00:06:30 All right. Speaker 00:06:31 The application of this is really as broad as the imagination Speaker 00:06:34 here are the core components. Speaker 00:06:35 The computer watches, the screen listens to the conversation, or just Speaker 00:06:39 plain old watches for events to happen. Speaker 00:06:42 In any number of administrative or clinical settings, and then it fires off Speaker 00:06:46 events, more code that makes something happen without human intervention. Speaker 00:06:51 And without human error. Speaker 00:06:53 At least once the process has been baked correctly. Speaker 00:06:56 You know, the old adage does apply here. Speaker 00:06:58 Nothing can mess things up quite as quickly or broadly Speaker 00:07:01 as a computer with bad code. Speaker 00:07:03 But if you bake it right. Speaker 00:07:05 You just relieved a burden of the clinician or the clinical team with a Speaker 00:07:10 process that may have taken days and may now be done in seconds and the Speaker 00:07:15 results passed on so that the next step in the process can fire off. Speaker 00:07:20 You know, this level of automation is being used extensively in billing Speaker 00:07:24 and administrative processes. Speaker 00:07:26 But it can really be applied to any administrative process in the hospital. Speaker 00:07:30 Take a high level event, like a patient schedules, an appointment. Speaker 00:07:34 Perhaps, there are things that need to happen before that appointment Speaker 00:07:37 around insurance check or information request back to the patient. Speaker 00:07:41 And many applications try to build that into the software, Speaker 00:07:44 but that was fairly rigid. Speaker 00:07:46 RPA sits alongside the applications and connects them in, in, in the best cases. Speaker 00:07:52 It connects them with API APIs, but in other cases with some Speaker 00:07:55 rudimentary processes to fire off additional processes, right? Speaker 00:08:00 So because it sits outside, it can be refactored many Speaker 00:08:04 times for different workflows. Speaker 00:08:06 Taken at the end of the process, the appointment is finished and Speaker 00:08:10 the claim needs to be created and submitted for payment. Speaker 00:08:13 It turns out that there are a ton of repetitive tasks. Speaker 00:08:17 That get done in the process and RPA is a natural fit. Speaker 00:08:21 To offload a lot of that burden. Speaker 00:08:23 With really high accuracy. Speaker 00:08:25 Rates, if not higher accuracy rates than than humans doing it. Speaker 00:08:29 Finally, let me talk about clinical automation for a second. Speaker 00:08:32 I recently interviewed the CEO for artist site, which is a startup Speaker 00:08:36 with a focus on clinical automation. Speaker 00:08:39 Here's what I learned. Speaker 00:08:40 The data in the EHR is not as clean as we would like. Speaker 00:08:43 And it's not always as real time as we would like in a clinical setting. Speaker 00:08:47 So they set out to solve the front end problem, which is Speaker 00:08:51 getting a set of clean data. Speaker 00:08:54 To act on and the simple solution is inexpensive camera in each room. Speaker 00:09:00 What can you do with a steady stream of data? Speaker 00:09:03 From a clinical room, whatever the various rooms in the hospital happened to be. Speaker 00:09:08 Well, you can look for fall risk, hand washing. Speaker 00:09:11 You can look at a pressure wound, patients being turned or moved, Speaker 00:09:16 and then you can send alerts to a nurse to move the patient. Speaker 00:09:19 Right. Speaker 00:09:20 But the reality is the sky's the limit because the backend Speaker 00:09:24 software can be trained to look for anything and fire off a process, Speaker 00:09:28 as simple as an alert to the nurse. Speaker 00:09:30 Or even more sophisticated with the use of API APIs and modern systems. Speaker 00:09:34 But, clearly the first thing that pops into your head is wait, how do clinicians Speaker 00:09:39 feel about being watched all the time? Speaker 00:09:40 And I'm glad you asked that question. Speaker 00:09:42 They aren't being watched all the time. Speaker 00:09:43 The code is trained to black out the identifying components Speaker 00:09:47 of the clinician, and just look for what is trained to look for. Speaker 00:09:51 So when you actually see the screen of what the computer is, Speaker 00:09:53 seeing it, blacks out those people. Speaker 00:09:56 So you don't really know who they are. Speaker 00:09:59 You might say, wait, I can probably correlate the data and the video feed Speaker 00:10:03 and figure out who was in the room. Speaker 00:10:06 You would, if the video feed actually existed, the system processes the Speaker 00:10:10 feed through AI and then forgets it. Speaker 00:10:14 Destroys the video feed and no longer needs it. Speaker 00:10:16 So it no longer exists quickly after the information is processed, it's gone Speaker 00:10:23 and the AI engine has done its work. Speaker 00:10:25 It is identified the things it needs to identify pretty interesting, right. Speaker 00:10:29 I was talking to one CIO. Speaker 00:10:29 Who's using this solution. Speaker 00:10:31 And their legal team came back and said, Hey, we need the Speaker 00:10:33 video feed from this room. Speaker 00:10:35 And I said, well, it doesn't exist. Speaker 00:10:36 And they said, oh, well, it has to exist. Speaker 00:10:37 There's a camera in the room. Speaker 00:10:39 So there's this underlying assumption that if there's a camera in the Speaker 00:10:42 room, there's a video feed somewhere. Speaker 00:10:43 So you have to educate people on what's actually happening. Speaker 00:10:45 But at the end of the day, the CIO was able to look at the legal team and say, Speaker 00:10:50 The video feed does not exist. Speaker 00:10:53 It does not exist anywhere. Speaker 00:10:54 It is gone. Speaker 00:10:55 It's not processed in that way. Speaker 00:10:56 It's not stored in that way. Speaker 00:10:58 So that's interesting. Speaker 00:10:58 It addresses that. Speaker 00:11:00 That that concern that big, brother's going to be looking over the shoulder Speaker 00:11:03 while adding an awful lot of value. Speaker 00:11:06 And even value to the clinicians in terms of alerts and those kinds of things. Speaker 00:11:10 And then, yeah, the obvious things like, Hey rooms empty, it needs to be clean. Speaker 00:11:14 You can get a quick feed into that room to see if the work has been Speaker 00:11:17 done and those kinds of things. Speaker 00:11:18 So awful. Speaker 00:11:19 Lot of value being created in that space. Speaker 00:11:23 Well, that's the world of automation. Speaker 00:11:25 It is big now and only going to get bigger in 2022. Speaker 00:11:30 So stay up to date on this one. Speaker 00:11:33 That's all for today. Speaker 00:11:33 If you know someone that might benefit from our channel, Speaker 00:11:35 please forward them a note. Speaker 00:11:36 They can subscribe on our website this week. Speaker 00:11:38 health.com/shows. Speaker 00:11:40 As we talked about earlier. Speaker 00:11:42 Or wherever you listen to podcasts, apple, Google, overcast, Spotify, Speaker 00:11:45 Stitcher, you get the picture. Speaker 00:11:46 We are everywhere. Speaker 00:11:48 We want to thank our channel sponsors who are investing in our mission to develop Speaker 00:11:51 the next generation of health leaders. Speaker 00:11:53 VMware Hill-Rom Starbridge advisors, McAfee and Aruba networks. Speaker 00:11:58 Thanks for listening.