The relationship between CIOs and clinicians is often complicated, particularly when veteran physicians are presented with quality metrics and asked to change the way they’ve done things for decades. In this interview, CIO Lee Carmen talks about the importance of collaborating with providers and making sure they don’t feel “attacked.” He also discusses what he’s doing to eliminate departmental IT purchases that are made without his approval, the benefits of centralizing core services, how he is navigating the best-of-breed versus enterprise-system issue, and the organization’s migration to Epic in all areas but the lab.
- Expanding Epic, but sticking with Cerner in lab
- The quest to banish interfaces
- HIMSS Analytics Stage 7 and device integration
- Voalte for clinician communications
- On leadership and staff management
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO SUBSCRIBE TO OUR iTUNES PODCAST FEED
All of our anesthesia machines are directly interfaced into Epic for data feeds. And we found, with some bumps in the road, pretty strong success in facilitating the direct flow of information into the clinical system through this, and it certainly improves or eliminates data entry time by our nurses in the ICUs.
We can analyze the data streams and we can do an independent validation to make sure that the firmware had absolutely no impact on the data stream. It’s created another layer of validation that is actually quite necessary that we hadn’t anticipated when we started down this road.
Our clinicians were wasting a lot of time trying to page somebody and then they wouldn’t be at the station that had the phone number they left. So our residents would be frustrated, our attendings would be frustrated, our nurses would be frustrated, and we ended up spending a lot of time messing around with phone tag.
We have people — mid-level managers, clinicians, and senior leaders — who have been working here in one capacity or another since the 60s. That’s recognized as a strength of our organization, and I think it speaks to the overarching culture and environment that we have here.
It’s a reward that we can go home at the end of the day and say we did something today that either helped somebody in a hospital bed or in a clinic, or it’s going to help some future cardiac patient.
Guerra: We talked about things being ready for primetime — the transplant module from Epic. You mentioned that you were still using Cerner’s lab system and I believe I’ve heard that a lot of people keep their lab system in place, even when they go to Epic. Tell me about the status of the lab system and whether indefinitely you’re going to hold on to Cerner, or what your plans are there.
Carmen: We’re actually running Cerner Classic, not Cerner’s latest generation of Millennium software, and through some required upgrades in partnership with Cerner, we can continue to run this until about the end of 2014. Then we need to either be transitioned to a more current version of Cerner’s software or to something else. Because of the nature of our contract with Epic, it could be a relatively simple task to move toward acquisition of Epic’s laboratory tool. Part of our installed Cerner application includes their Blood Bank Management System, and Epic has chosen up to now not to do blood bank. So right now we’re about to put an RFP on the street for a blood bank system and evaluate that. And we’re actually closely evaluating Epic’s laboratory system, Beaker, to determine whether it’s really where it needs to be for a large, busy, academic medical center’s laboratory environment.
Should that be the outcome of that review — and we expect to have that concluded in a couple of weeks — we will very likely proceed with implementation of Epic’s lab system and a third-party blood bank system. But again, building in on the same strategy; having all the systems in the same database has proven to be very beneficial to us, and we think it will continue to be. With some of the workflows that we see associated with gathering pathology specimens in the OR and transferring to laboratory, we believe having a laboratory environment in the same environment as the rest of our systems could really positively impact those workflows. So we’re really excited of what the potential might be with that.
Guerra: Anything that gets another interface out of your hair is a good thing.
Carmen: Absolutely. It’s actually twofold. It’s the operational interface and trying to eliminate that potential for disruption or delay, and on the backend, it’s the reporting side and being able to improve our analytics capabilities by eliminating the remapping of data dictionaries amongst disparate systems. I burn a lot of staff time and a lot of complexity in those settings.
Earlier in this discussion, we talked about performance metrics and physicians questioning the metrics. Often when we’re trying to interface different systems and merge disparate data, they have good reason to question the data. But in our opinion, moving more and more to a more integrated data model puts us on a better footing for those kinds of performance metrics.
Guerra: You talked before about device integration. By the way, congratulations on being awarded HIMSS Analytics Stage 7.
Carmen: Thank you. We’re very excited about that.
Guerra: The quote from John Hoyt of HIMSS talked about the integration of all devices in the ICU and emergency room so that transcription error is eliminated in recording values from devices, including blood administration and vital patient information. So you talked about that; tell the readers and the listeners a little bit about what you’re doing around device integration.
Carmen: The technical architecture that we have in place — Epic uses a third party company called Capsule that acts as the interface gateway between medical devices and the Epic system. We’ve heavily leveraged the Capsule technology across our inpatient settings, the ED as you mentioned, and also through the OR, we’re running Epic’s Intra-Op Anesthesia module. So all of our anesthesia machines are directly interfaced into Epic for data feeds. And we found, with some bumps in the road, pretty strong success in facilitating the direct flow of information into the clinical system through this, and it certainly improves or eliminates data entry time by our nurses in the ICUs. We did one brief study that suggested on a 12-hour nursing shift in the ICUs, they were saving about 55 minutes of time just with data entry.
It certainly improves the accuracy of the data in that someone’s not accidentally mis-transcribing some data element. And it provides our physicians access to that data wherever they are as the data is collected. If for some reason the physician is at home or in another part of the enterprise, they can log in and they can see the real-time data. Our intensivists use that pretty heavily to keep track of patients that they’re very concerned about.
It drives other behavior in the organization. What we didn’t realize at the time is when some of our bedside device vendors wanted to upgrade the firmware code on their devices, they would either come on site or remotely do that, and it largely was an unnoticed event by our care providers and by our bioengineering staff. They just came in and they would flash the firmware up to the latest version and they would be gone and nobody really had to do anything about it.
What we found though, as we move forward with device integration, is that occasionally the firmware upgrades would change the formatting of the data that was being streamed out of the system to Epic. That was an impact that we never experienced before because we weren’t doing anything with the data feeds. We were just looking at the console of the device. But we accidentally found cases where suddenly data was streaming out of a device that made absolutely no sense. It was being truncated improperly, and so it was populating the record improperly. The nursing staff saw it immediately and called it to our attention, and when we analyzed the problem, we found that the vendor had just flashed the firmware on that device.
And so it caused us to actually rework the entire change management strategy for maintaining software on the biomedical devices so that before anything gets flashed to a production device, it goes through our simulation lab and we can actually do the upgrade. We can analyze the data streams and we can do an independent validation to make sure that the firmware had absolutely no impact on the data stream. It’s created another layer of validation that is actually quite necessary that we hadn’t anticipated when we started down this road. But we found it really necessary to ensure the validity of the data stream.
Guerra: Tell me a little bit about what you’re doing around communications with Voalte and how that’s helping your overall flow and communication between the clinicians?
Carmen: I’m glad you asked about that. That’s a really exciting project that we have underway. We have been running in our organization a proprietary digital paging system for a very long time. Because of where we’re located geographically and the fact that we’re not in a major urban area, we have found it impossible to partner with a communications vendor for paging services. They just cannot or will not agree to a service level agreement that gives us the kind of response times we need for certain emergent situations. If we have a code blue team, our cardiac response team carrying the pagers, they’re expecting paging throughput of 10 seconds or less, and we found when trying to negotiate with commercial carriers that the best they would agree to is about 20 minutes response time.
Guerra: That’s a big difference.
Carmen: That’s a huge difference. So we continued the burden of running our own paging system, but as a result, we probably didn’t progress as quickly as some of our partners in a more urban setting with technologies such as bidirectional texting. And like many large facilities, we don’t really have very good cellular coverage within the enterprise. So what we’ve decided to do is, as we built out our wireless infrastructure to support a lot of mobile computing, we then started looking at RFID and we partnered with AeroScout and we started employing RFID. By then we had constructed a pretty dense wireless network that was quite reliable, and so then we started moving to voiceover IP with mobile handsets. And while our clinicians and our staff really liked the flexibility of having that mobile telephony, the devices that were available — and to some extent, are still available — are pretty inadequate.
Most of the Cisco or some of the other handsets really perform like a cellphone from the mid-90s. What I routinely hear is these bricks. Why are you giving me these bricks to carry around for? And so when we became aware of Voalte, who was clearly doing some very interesting things with what I refer to as neutered iPhones — they don’t work on the cellular network; they are engineered and constructed to only work on our wireless network — we did a small pilot with about 65 phones, and the results were just overwhelmingly positive. Our clinicians, both nurses and physicians, and including social workers and pharmacists, were wasting a lot of time trying to page somebody and then they wouldn’t be at the station that had the phone number they left. So our residents would be frustrated, our attendings would be frustrated, our nurses would be frustrated, and we ended up spending a lot of time messing around with phone tag.
We eliminated a lot of that with the Voalte system on this one inpatient unit. That was before we even got to things like nurse call integration or cardiac monitoring integration. We’re working right now to roll out about 1,200 iPhones, again with the basic functionality of telephony and bidirectional texting, and with some additional apps to support medical knowledge — reference libraries and those sorts of things. We are validating right now that Epic’s next generation of their mobile client can coexist peacefully with the Voalte application and testing suggests it can.
And we seem to have a lot of interest between Voalte and Epic in working together. For example, the Voalte tool can read from Epic who’s on the care team for a patient, and can facilitate the staff member to either call or send a text to everybody on that care team to update them on the status of a patient. We’re also working on our nurse call integration right now. We’re really excited. Among all the projects I have underway right now — and I have a lot of projects underway — that one the physicians in particular are extremely aggressive in trying to shorten the rollout of the system. But we’re very excited. Voalte’s been a great partner to work with so far. We think the platform has a lot of capabilities that we’ll be able to build on to get our providers to the point where they only need to carry around one device and do an awful lot of things with that one device.
Guerra: We’ve got about a few minutes left. I’m not going to break my rule of keeping people more than an hour, and you’ve been very generous with your time. I just want to talk a little bit about leadership. I was reading some interviews you’ve done in the past and you talked about the longevity of the team you’ve got. You’ve mentioned a number of times during this interview what an excellent team you have. You said two of your directors have over 30 years of experience each. Do you have any management philosophy that you think really promotes people sticking around? Is there anything about how you treat your employees or people on your team — anything you could share in that area?
Carmen: Well, in full disclosure, I have to say that our organization seems to be very lucky and very unique in that we tend to have very long tenures of service in all areas of our organization. We’re not sitting in downtown Chicago where we have other medical centers blocks away that someone could leave our organization and go to. We have people — mid-level managers, clinicians, and senior leaders — who have been working here in one capacity or another since the 60s. That’s recognized as a strength of our organization, and I think it speaks to the overarching culture and environment that we have here.
In my particular area, I have two goals for the IT enterprise. Obviously, the first goal is to deliver the best service we can in an affordable, efficient way, but the second goal is really to support a positive work environment for the IT staff. Everybody recognizes that there are days when you go get up in the morning and it’s cold and it’s raining you just think, I’d rather go back to bed. For people, day after day they get up and come in to work, I’m lucky that I have a number of staff who actually work for us acknowledging that they might receive or probably would receive greater compensation working for somebody else in the region. But they come to us because either as a child they experienced some major medical event and were treated here, or their siblings had some major medical event and were treated here, or a family member had some medical major event and were treated here. And as a result of that experience, they have a great passion for our organization and feel a strong sense of purpose working for a healthcare delivery organization.
I don’t mean to demean other industries, but if you work in manufacturing, your business is to create whatever widget you’re making, and that’s your goal. My team here, we’re all housed onsite, and so every day we walk to the halls and we pass toddlers who are dealing with cancer or dealing with transplant or dealing with trauma who are being wheeled through the halls or are experiencing very extensive surgeries up in our ORs or our cath labs. My team, I think we’ve been very successful in driving home to my staff that they’re not just IT people. They are part of the care delivery team.
As I mentioned at the beginning we support the three missions of the organization, the academic, the research and the healthcare delivery. Even the staff whose day-in and day-out duties don’t touch the healthcare delivery part of our mission but are more engaged with the academic mission or more engaged with the research mission, we’ve been able to get them to understand that by supporting the medical school, they’re impacting the delivery of healthcare for six to eight years from now. And so it may not be the patient in the OR today, but it will be a patient in the OR in the future that they’re having a positive impact on. Likewise, in the research environment, they may be supporting the guy who’s going to cure cystic fibrosis or going to find a cure or a treatment for macular degeneration. Again, maybe that’s not going to happen tomorrow or maybe that’s not going to happen two years from now, but maybe 10 years from now they’re going to be responsible for curing a whole segment of our population.
That’s a theme that we drive home and it’s sort of double-bladed in that there’s a responsibility to that. There’s a service level we strive to achieve so that we’re not a barrier to supporting these services. And in fact, we enhance our staff’s ability to deliver these services. But it’s also a reward that we can go home at the end of the day and say we did something today that either helped somebody in a hospital bed or in a clinic, or it’s going to help some future cardiac patient.
Guerra: Well Lee, that’s fantastic, and like I said, I’m going to be sensitive to your time. We could go for another half hour, but let’s leave it there for today. Thank you so much for your time.
Carmen: It was a pleasure talking to you.
Guerra: All right, I’ll be in touch. Have a great day.
Carmen: You as well. Take care.