Dennis Loftus, Senior VP and CIO at Riverside Health System in Virginia, is one of the four participants in a symposium Gevity and U.S. partner Divurgent are hosting at eHealth 2019 on May 28 from 11:15 a.m.-12:15 p.m. The event, which focuses on how healthcare organizations can maximize their Epic implementations, will share lessons learned and strategies for success.
Dennis spoke to us about his role at Riverside and his experience implementing Epic over the past few years.
Gevity: You’ve been at Riverside for a long time. What keeps you there?
DL: I joined Riverside, a not-for-profit healthcare organization, in 1992. Riverside has five acute care hospitals, three speciality hospitals, a large medical group, a dozen nursing homes and various other healthcare related services. What’s kept me at Riverside are: the people, the mission, the excitement and challenges in the information technology field and knowing the purpose of the organization is to take care of the communities we serve. I’m winding down my career now so I will be retiring soon, although I will probably do some interim CIO work here and there. I feel as though I’ve worked with just about every healthcare computer system that’s out there!
Gevity: Riverside implemented Epic in 2017 to consolidate a number of previously existing systems. Were there any other drivers and goals? Were those goals realized?
DL: Over the years, we had taken a best of breed approach at Riverside which resulted in different electronic medical records (EMRs) for our physician practices, emergency department, acute care hospitals, rehab facility, behavioural health and nursing homes, and it was becoming too difficult in so many ways. The more disparate systems you have, the more you risk issues with patient care, safety and experience. You also have more technology issues as you try to keep them all in sync. For example, if we upgraded one EMR, invariably we would find out that it would only work with a certain version of a web browser, but that web browser version did not work well with one of the other EMRs, or one of our add-ons, such as anti-virus software, would not work with the upgrade.
Trying to integrate all of this internally through a health information exchange (HIE) or a context linking tool that is designed to link patients from one system to the other became too complex and too frustrating. Theoretically, it sounded good, though: we implemented context linking for our emergency department (ED) docs, and the idea was that when they were in their EMR and they wanted to look up the record for a patient coming from the acute care hospital or physician practice, the software would find that patient and log the doctor into that application where they could then find the patient’s information. But it took time and the physicians don’t like that; as well, they had to learn to navigate a totally different system. They got frustrated; they liked the concept, but after three to five months, they stopped using it because it was too difficult to use and it slowed them down.
We also found that some of the vendors we were using were having difficulties keeping up with and meeting some regulatory requirements such as Meaningful Use. From about 2014 to 2017, many of those companies were going out of business or faltering, so we decided to look at the three main EHR solutions – Epic, Cerner and Meditech. We did our due diligence and realized Epic was the right solution for us. In addition to patient care benefits, we also wanted to realize revenue cycle improvements and operational efficiencies.
A large complex healthcare organization should expect the lifespan of a large EMR to extend beyond 15-20 years. So, while there is an initial significant cash outflow, Riverside’s financial break-even analysis showed a six-year payback: we’re saving about $6 million a year by having switched to Epic and eliminating disparate systems. You hear these horror stories of how costly these systems are, but we’ve actually driven costs out of our system and improved the revenue cash inflows, so this has been a huge success for us. Although probably no physician likes an EMR, our docs are saying we have right strategy, so all our goals were either met or exceeded.
Gevity: Why do you think your experience was so positive compared to a lot of others?
DL: Riverside had invested in different EMRs over the years, so we had developed a strong staff. They had a lot of experience, not necessarily with Epic, but with the concept. We also had a pretty stable IT team and organization to begin with. I also think a lot of organizations that embark on something like this spend a lot of money on consultants when they probably don’t need to because they could be developing a lot of internal experts – that’s what we did.
We did use external consultants here and there – we used Divurgent to provide go live support, and they did a fantastic job – and we used other vendors to plug gaps or weak areas, but we were very careful about that.
Gevity: Can you describe how implementing Epic contributes to improved patient safety?
DL: When a patient goes from one setting to another or has a significant transition of care, you have to reconcile their medications. If I as a patient go to see my family doctor and the doctor reconciles my meds at that visit, that’s great. But then if I injure myself a week later and go to the ED, where they are using a different EMR, while they also have to reconcile my meds, they’re not starting with the most current data. If I have to be admitted to the hospital, which also uses a different EMR, they, too, have to reconcile my meds, but again, they aren’t using the most current data. Every hand-off or care transition is one more risk area for a patient, so by having one integrated system, the most up-to-date information is always available to caregivers.
Another example is that Epic has done a pretty good job of collecting evidence-based best practice protocols and built it into their software. Say, for example, you want to know the best practice protocols for certain procedures; through Epic, you can get the protocols perhaps that Johns Hopkins, Duke, Stanford or others used. With this as the foundation, we could present this to our clinicians and then modify them as appropriate.
Gevity: Were there any big surprises that you experienced in your implementation?
DL: I was surprised by how stressful and time-consuming and tedious it was to form the team. There were a lot of HR and compensation issues and it was a lot more work than I expected to figure out who to pull from IT and operations and then backfill support for legacy systems. We also underestimated go live support requirements, so we didn’t budget as much as we should have and we didn’t have buy-in from senior leadership because I don’t think they understood how important that was going to be until maybe three to six months before our first go live.
Another big surprise was understanding the importance of how to keep the team together after you go live; you have just invested two years in your resources and spent money on training and certification, but all of a sudden, those resources become very sought after and there are a lot of consulting firms and other organizations that would like to steal them from you!
Gevity: Did you engage clinicians in the process, and if so, how? What was the impact of that engagement?
DL: Absolutely! That is so important. In our case, we had both our chief medical information officer (CMIO) and our associate CMIO entirely dedicated to this work. We also had committed, supportive service line chiefs responsible for different service lines. If possible, any organization implementing Epic should invest as much as they can in physician builders – doctors who receive focused EHR training – because those physicians can save you in so many ways, including ensuring the system is built correctly and helping to educate and get buy-in from other physicians.
Gevity: Are there any lessons you learned that you think are relevant to the Canadian context?
DL: With Canada’s decentralization of health care, integrating disparate systems will be a challenge. HIE solutions will help but can also provide many technical challenges. In Virginia, for example, nearly all of the large integrated healthcare systems use Epic. This has made it incredibly easy and seamless to share health records with nearly all of the other organizations in the state. Since implementing Epic less than two years ago, Riverside has shared over three million health records with other organizations, including all 50 states.
Other lessons I believe are applicable are:
- It is critical to involve physicians, nurses, pharmacists, and all operations areas of an organization during the requirements definition, build and implementation phases.
- Invest in your internal people as much as possible and fill the gaps with outside expertise. Encourage the use of physician “builders.”
- Be thrifty, judicious and careful about spending money early in the process.
- Be willing to invest in “at the elbow” support during the go-live phase but monitor these costs closely.
Also, keep in mind that when you are given a project and your board signs off on a $100 million or $200 million project, you think you’ve got all this money, but from Day 1, you have to be very disciplined about where you spend that money, because it goes pretty quickly. All of a sudden you find yourself toward the end, and you could end up over budget. In our case, we were under budget by quite a bit, so we did pretty well.