Migrating Epic to Azure Starting With Why

Transcript

Derek: Welcome, everyone. Thank you for joining our panel discussion, “Migrating Epic to Azure, Starting With Why.” On our panel today we have two experts in cloud migration and hosting from Tegria. Tegria is a global healthcare consulting and services company delivering solutions that leverage technology to help provider and payer organizations transform healthcare. And Tegria is one of Microsoft’s Select Epic on Azure preferred partners. We’re happy to partner with Microsoft for this discussion today.

I’m Derek Garrison. I’m the Microsoft Channel Partner Manager here at Tegria, and I’ll be the moderator for our panel discussion.

Sam: I’m Sam Coker. I’m the Global Black Belt for Healthcare specializing in Epic. Thank you for having me.

Tom: My name is Tom Miller. I’m the Director of Systems Management and Monitoring Services here at Tegria. I’ve been working in the Epic space, starting at a provider since 2006 as a part of and leading the technical teams that manage Epic infrastructure.

Chad: I am Chad Skidmore. I’m the Vice President of our Technology Services area, which covers a lot of our cloud work for our healthcare customers. I’ve been in this role for the last 19 years.

Derek: Thanks, all right. The EMR is the crown jewel of the IT estate for a healthcare organization, and the decision to migrate it to the cloud is important. It can be complex, and it can represent a risk. And, we find that this decision is different from other decisions IT groups have made in the past.

In the planning stages for deciding whether to go with an EMR or an ERP, how you answered the question “why?” was a little more straightforward. And, as a result, the “how” of the implementation methodology (the project management, the budget) got more attention. But we find in our experience working with customers to migrate the EHR to the cloud, how you answer the question, “why?” (your motivations, the benefits you’re expecting to see) can point you to different paths forward. So, giving this question due diligence—it’s critical early on—is the topic for our panel discussion today.

To get us started, I’d like to ask our panelists: In your conversations with customers in the early stages of planning a migration, what motivations are you hearing most, and what are your thoughts on what you’re hearing?

Chad: I think one that we hear often is “we just want to get out of the data center business.” And a lot of times, that might be driven by a myriad of different things that are in front of the customer, in front of their business, at the time. That might be they’re running out of space within the current data center as they’ve grown, added more facilities, more outpatient clinics, etc. Or they might be out of critical infrastructure items like power and cooling, etc. Maybe they are facing a big capital spend for a hardware refresh and now is the appropriate time to evaluate moving out of that model and into more of an OpEx style of model like we would see in a public cloud. So, it could be a number of different things that I sum up as “we want to get out of that legacy data center kind of a business.”

Tom: I’d say somewhat related to that is some staffing challenges because you may have staff that are highly focused on keeping that data center up and running, managing day to day, keeping the lights on, that sort of thing. And you also have the organization that has a lot of projects that they want to get going that heavily involve IT that improve the patient experience, improve the end user experience, that sort of thing. And, due to your staffing, you don’t have time for those staff members to focus on those projects because you’re focused on keeping a data center up and running. So, you look at how do we shift that out so that we’re not managing that data center, and we can reallocate those resources to help accelerate the goal of these other projects.

Another factor is looking at rapidly responding to changes that may affect your business. We had a pandemic a few years ago and a lot of healthcare organizations are suddenly having to schedule a whole bunch of testing, a whole bunch of vaccinations, etc., and that put additional load on their infrastructure that’s running their EHR. In the traditional model, you would have to spend months and money and time to get all that infrastructure in place so that you can meet those demands. With the cloud, you can more rapidly deploy that additional infrastructure that you need, and then once that demand lowers, you reduce that cost by then turning those systems off. Whereas, in the classic model, you’ve already spent that capital.

Sam: I’m hearing the same things Tom and Chad are hearing. And the only other thing that I would add to that that I’m also hearing is a lot of times the business is driving this decision. So, the business is coming to our department (IT departments) and saying, “Hey, we’re moving to the cloud. How do we get there?”

Tom: I would say one other thing is cybersecurity. Lots of stuff in the news lately about breaches and ransomware attacks. Organizations are looking to the cloud to create a more secure environment for their infrastructure. Maybe it’s just their disaster recovery (DR) environment, but it creates that isolated instance of their EHR so they’re better protected from ransomware and other types of attacks.

Derek: Also in our conversations that dig into “why,” I know that sometimes we find that the motivations our customers have can be misdirected. So, for our next question to the panelists, I’d like to ask, can you give some examples of this that you’ve encountered?

Chad: I think some either misdirected or just some things that folks overlook is that getting out of the data center business, as I mentioned earlier, really just means getting out of the data center business. It’s servers someplace else. It’s infrastructure someplace else. You still have to manage that, care and feed it, patch it, repair it when it breaks, and those kinds of things. So, you’re not completely getting out of that overall backend technology business, you’re just moving it around. That’s not to say that it’s not a good thing to do, but just don’t discount the fact that you still have to maintain all of those things.

The other thing that I run into sometimes is clients not fully understanding what that total cost of ownership model is going to look like. They assume or believe a lot of the sales pitches and the hype that says it’s going to be way cheaper. That may not necessarily be true. I think certainly longer term it can be; but right out of the chute, you may not be saving money. You may have a higher spend for a while, up until you optimize and start to really embrace and take advantage of some of the benefit of moving to the cloud that does drive that additional savings.

Sam: I just really want to reinforce what Chad’s saying on the “whys.” I see two big ones. One is budgetary discipline that you really need to have on these environments. Moving to the cloud, you need to understand, and you need to tag your environments, and you need to look at it as a proactive way, not as a reactive way. So, look at what your spend is going to be next month instead of what your spend was last month.

And then along those lines is that operational discipline that we need, even more so in the cloud. As we move, the big benefit of the cloud is moving to code, moving to operationalizing or coding our operations to kind of remove the human factor, and also to automate and enhance. That requires discipline, staffing, and skill enhancements.

Tom: To add on to discussion of our “whys” here … you’re looking at budgetary, you’re looking at operational. You also need to consider the skill sets of your staff because it’s going to be different sets of skills.

You may be looking at it, as Chad mentioned earlier, as another person’s computer, or as computers in another data center; but it’s a whole different management system for managing that infrastructure. VMware is vastly different from Azure when you look at managing the resources and the infrastructure. So, making sure that your staff is properly skilled and you’re working with a partner that can help you accelerate that process is key to being successful.

Derek: So the first two questions were around what we’re hearing from customers. I’d like to turn that around a little bit and next ask what are the things that you’re not necessarily hearing from customers, but you think are important for them to consider?

Chad: I think one of the things I don’t hear very often is that it can be one step toward really improving the overall security posture. In today’s day and age, we see frequent cyberattacks on healthcare organizations. It’s a real and present danger in this space today. By moving some of those core IT assets into the cloud where they can be walled off better (moved away from that user environment), additional tool sets that are not available or are harder to attain in an on-prem environment can be deployed to better protect the core IT assets.

Tom: Another thing that we’re not necessarily hearing about from customers is data and analytics. A lot of times, that’s an isolated work stream; but when you think about getting your EHR out into the cloud, you’re now moving all that data into the cloud, which is much more powerful, robust, and flexible. You are taking that data and converting it into information that can be consumed by senior leadership for making organizational decisions. Having that data in the cloud accelerates the implementation timeline to realize the benefits of getting that data turned into information that can be consumed.

Chad: Another thing that I don’t hear frequently is “we’ve got a really good understanding of our current state.” That can mean a lot of things. Total cost of ownership might be one; fully understanding and rationalizing your current state total cost ownership is difficult at times. You may not be including all of your data center costs or all of your backup power in an infrastructure cost because maybe they’re shared with other things within the healthcare organization. When you’re comparing against the cloud, that can skew that total cost ownership comparison.

Look at staffing costs and training costs required to move into the cloud environment, ensuring that those are rolled into an overall plan and analysis. Also take the opportunity to go through an app rationalization exercise and really question, “Why do we have three or four apps that do very similar things? Is this an opportunity to consolidate, to shrink our IT estate overall?” Consider dropping some legacy applications that are not cloud-ready or cloud-native in favor of those that are to allow for better optimization once you do land in that public cloud.

Sam: I want to pull on this thread a little bit more because some people have this idea that they will move to the cloud and it’s all going to be happy and done. The challenge is you’re going have a long period of a hybrid design where you’re going to have on-prem and in the cloud. What I don’t see discussed is “how do we take the lessons we’re learning and rationalizations were making in the cloud and move them back in to improve our on-prem as well?” That way, we have a unified painted glass across the spectrum as we’re running in this two-legged system and treating this as one unified system that we run our applications on.

The other thing that I think is very related is we need to ask ourselves, “why are we moving to the cloud, and what are we going to get from the cloud?” That’s the real value. Once you move your data into the cloud, you open up a whole world of interoperability, of business intelligence tools, of AI tools, and you’re prepping yourself technically for the future.

So you need to ask yourself, “At the end of the day, the apps that I’m running today are going to run the same way on the cloud (I still need to patch them, and I still need to update them). But once I get them there, what can I do to improve and to lower cost?”

Derek: At the beginning, I mentioned the importance we found of clearly articulating “why.” What are your motivations to move your EMR to the cloud? Your answers to these questions are going to direct and shape your strategy. Next, I’d like to ask our panelists for examples of how they’ve seen the answers to these questions connect to strategy.

Chad: Sometimes I think this is the hard part. Sometimes the “analysis paralysis” gets in the way of just taking a leap and getting started. I don’t want that to sound too cavalier, but at some point you need to pick a starting point and jump in to start to learn a little bit more about those things that you don’t know. Prove out some of your hypotheses or aspects of your plan. Look at test-trained, non-prod workloads. You will, by moving those to the cloud at relatively low risk, start to understand things like “Where are our skill sets a little deficient? Where do we have gaps? What are some of the hesitations overall?” We can overcome those in that safe environment. And I think the app rationalization exercise that I mentioned earlier can also be an opportunity to find early adopter applications or application sets to move. Maybe you’ve got some legacy applications that you’ve identified are a great opportunity to go cloud-native. Do a pilot project with something like that. That’s also a fairly low-impact, low-cost way to approach.

Tom: When you’re thinking about your migration strategy, specifically Epic, you can ease into it with a low-cost, proof-of-concept type of thing. Training is a great example. You could put one or two classroom environments up there and get some real-world, end user experience with actual workflows of the students that are in the classroom as opposed to starting with a POC or TST environment. Then, all you’re getting is the experience of your technical staff that are doing the Epic build and the technical configuration, etc. Training allows you to ease in with a much smaller footprint of one or two training environments. You can refresh those from your on-prem, or maybe you put a prep environment up there and use your environment refresh process to more frequently refresh those classrooms to get even more experience.

When you’re looking at migration strategy, if cybersecurity is your primary concern, DR is probably the best environment to start with (when you’re looking at Epic). That allows you to create a safer starting point. You’re not looking at production—you’re not putting your production at risk. You’re creating that isolated environment that is going to be in a different data center, geographically diverse from where your production currently is. That allows you to ease into that process and validate that it’s going to work for you. You can do some test runs to validate performance and that sort of thing.

Sam: I’m really with you on that one. I think how we connect this to strategy is, as we mentioned earlier, you’re going to be hybrid for a while. And I like what Chad just said, the tech’s not the hard part. Trust yourselves. You all know how to run the technology. You know your environments. But really where you want to spend your time and focus is around governance and the operational part. Lean into the cloud to do the operational part to standardize, to run scripts, to create environments that are unified and single panes of glass. Every bit of time you spend there is going to be value, not just for this specific application move, but for your whole journey on the cloud.

Derek: Excellent, thank you. To wrap up our questions, I’d like to ask each of our panelists for any parting thoughts or advice.

Tom: It’s important to understand that this is going to require different skill sets to build and manage the infrastructure because you’re not just looking at servers. Now you’re looking at your network, your firewall, everything in one place that you need to manage. Make sure you’re thinking about that and look at including a trusted partner to help you accelerate that implementation. Also, if you’ve got a partner that has good talent for managing your Epic infrastructure and other infrastructure, you can have them help with that while your team spends more time focused on your migration plans.

Chad: I think I’d sum it up by saying be ready to embrace change. For those early POC opportunities, choose the folks on your staff who are most open and accepting to letting go of the way they’ve always done things and embracing change. Get somebody who is ready to do infrastructure as code and embrace orchestration. That will break some people’s brains a little bit if they’re very old school IT data center operations folks, but that’s how you’re going to really start to take advantage of and see some value from adopting cloud.

Sam: The best advice I can give you is don’t do this without a partner. You really need to learn from others’ experience. I’m not saying the partners do everything. What I’m saying is look for a partner that will teach you how to do this better, how to get your run books up to speed, and how to use infrastructure as code. I started working on Epic on VMS, and if I can learn, anybody can learn. Learn from others and seek out advice—the successful ones are going to ask for help.

Derek: Excellent. I’d like to thank our panelists. I think you brought up some interesting and salient points today. On a final note, if you are considering migrating your Epic system to the cloud, this QR code will take you to Tegria’s webpage on moving your EHR to Azure, and there you’ll find a form to contact us. We would love to discuss your cloud migration with you, working through your motivations, whether you’re in the planning stage or at any point in that journey. Thank you.