HIStalk Practice Interviews ChenMed Chief Quality Officer Jessica Chen, MD and CTO Denise Hatzidakis
Jessica Chen, MD is chief quality officer and Denise Hatzidakis is CTO at ChenMed.
Tell me about yourself and the company.
Jessica Chen: I’m chief quality officer for the ChenMed family of companies. I’m an internal medicine physician with a strong background in quality. I cover all of our quality metrics, as well as our onsite medication protocols and systems. I work very closely with our IT team to ensure that our technology always has a physician focus, and we try to make it as efficient and effective as possible for enhancing the clinical encounter.
We call ChenMed a family of companies because we have a few different seats under it. Altogether, we have 39 medical centers across six states. We call it nine different markets because within those states we have different locations or different cities that we cover. Our centers are focused on seniors primarily in under-served areas, and we focus on Medicare Advantage, where we feel we can have the best impact on the full spectrum of care. We’re a totally capitated, at-risk practice, which means you really cover end-to-end care – from inpatient to outpatient, we’re responsible for it all. It allows us, as a clinical organization of physicians, to really take ownership of a patient’s total care. Our goal is to align our patient’s outcome with our outcomes.
Denise Hatzidakis: I’m the CTO for ChenMed. My role is building. I’m responsible for all the technology systems that support our physicians and our care model in other parts of the organization to make the practices work. I have a software development team of 50. I grew up as a "Will Code for Food" computer geek and have grown into really loving, and enjoying maturing and building IT systems. ChenMed gives me the opportunity to do that in a space that really makes a difference.
Jessica, you’ve been with ChenMed since 2009. How have you seen ChenMed’s use of technology evolve alongside the evolution of federal programs like Meaningful Use?
JC: We actually moved to an EHR in the early 2000s. It was non-certified at the time; there wasn’t really such a thing as certification. We had been an early adopter of technology all along. It was a hybrid system that was still somewhat paper-based, somewhat digital. When Meaningful Use came out, we did look at other technology systems. We looked at all the incentives and, obviously, the penalties, but we also had to balance that with our clinical model and what we felt was needed from a patient care standpoint. When we looked at the standards and we looked at the measures … because we’re very focused on seniors, not all of them necessarily applied to our population. There were not any IT systems at the time that were built for value-based care. We had to make a decision: Do we care more about checking the box of Meaningful Use or do we care more about our technology working in a value-based care system?
We made the decision to build something because there was nothing out there that was built for value-based care. We actually did initially look at going for certification, but realized it actually didn’t make sense for us. The majority of the Meaningful Use incentives really targeted fee-for-service medicine. We have very little fee for service. We’re really focused on Medicare Advantage, and so the incentives just didn’t work the same in that population. When we make decisions around what’s prioritized in our development and what we should put in for our own technology, we do it based on what’s going to improve outcome. What’s going to improve our physician’s efficiency? What’s really needed in our practice?
How did implementation and adoption go with that home-grown technology?
JC: It’s been a little bit piece by piece. What we developed five years ago is very different than what we have today. It’s pretty much been designed by physicians. Denise’s team built it. We take in a lot of feedback from our physicians on the ground, from all the markets. We get a lot of requests and then we kind of conglomerate all of that and ask ourselves, what’s really going to be the highest priority for them and what’s going to impact the business the most? What will improve clinical care, decrease hospitalizations? What’s going to help us do all of this best – that’s what we put in there.
Because we’re typically targeting things that the physicians have asked for, our adoption has been pretty good. We’ve had physicians come out of other systems and it takes them a little bit of time to get used to it, but then the say, "Oh, this is so much easier to write a note than clicking 800 times." The template doesn’t have all the bells and whistles that some of the certified EHRs have, but it really targets our model and so I think that’s really helped adoption.
DH: We use a user-centered design approach to design our software, so it really is technology for physicians designed by physicians. We sit side by side with all of our care providers to understand how they do their job. We have tools that surround our EHR and the data in it that are specifically targeted for those different roles, so the design is specifically for the PCP, or the medical assistant, or someone interacting with our system outside the clinical setting. It’s designed by the users, so to Jessica’s point, it makes the adoption occur much more quickly.
Since adoption has gone so well, have you noticed your end-users becoming more sophisticated in their expectations of what the technology can do? Have you set too high a bar for yourselves?
DH: That’s kind of normal in software development, right? Once you get the "must haves" out of the way – once your end users realize you have the capability to do bigger and better things – then yeah, the requirements keep coming. That said, I defer to Jessica as to hitting the mark with respect to the functionality they have now and the balance between that and what they’d still like. From the IT perspective, we work very, very closely with our end users. We meet with them weekly, and there’s always someone who asks, "Oh, if you could do this, could you do this too?" That’s very normal when you have user-engaged software development.
JC: Our physicians really do feel accountable for their panel, and so they’re constantly saying, "Hey, if you could do this that or the other for me, it would make me more efficient or more effective, or I could prevent hospitalizations if I could have this information." We have non-conventional things in our EHR. For example, our physicians have requested to know how many of their patients are in the hospital at any given time. There’s no system out there that does that, and most systems are not interoperable with hospitals. In most care settings, you have no idea what is happening with your patient outside of your center. And so we’ve had to build our own technology to be able to capture that data. We know when they’re in the hospital. We follow them closely while they’re there. We know when they’re getting discharged. We can go see them in the home. We want to get them back into our center. We know if they’re in assist.
We try to really coordinate that care and the doctors want to see all that. They tell me, "If you want me to be responsible for improving our hospitalization rate, then I need more information." We’ve created a lot of technology around that, but then again, there’s constantly new ideas coming in. Expectations, I think, are always increasing, but we definitely see that as an opportunity to build really neat and innovative tools.
How have you seen this technology affect outcomes?
JC: We’ve definitely seen a clear drop in both ER visits and admits, and hospital sick days because we’re getting the patients in to see us, and we’re trying to do more on the preventive side. A lot of that, I think, is driven by just knowing what’s going on with the patient and being able to identify where they are – to help get them into the right level of care and provide more access to them.
As you’ve been developing this technology over the last several years, have you looked to any other providers or vendors for inspiration or advice?
DH: I think we’re always looking at the state of the art in the industry and trying to understand what makes sense for our care model. What is applicable, and is it something that we should build in because it does provide better outcomes? We spent a good bit of time at HIMSS looking at what people are doing in value-based care. What does is it mean? It means a lot of different things to a lot of different people.
What’s next for ChenMed when it comes to building or buying technology? What’s on your wish list?
JC: We’re always going to be innovating and creating things internally, for sure. That’s never going to stop. I think one of the things we’re trying to tackle, just like everybody else, is interoperability and trying to get more data from more sources. Our experience has been, the more information we have, the better. We’ve been trying to get real-time ER notifications. If our patient hits the ER, we want to know about it and we want to make sure we follow up. That’s actually been one of the hardest to get for a number of reasons. We are already getting more real-time hospital notifications, but it varies somewhat based on the hospital.
I think our biggest ask would be better interoperability and having the full spectrum of care – from hospitals to nursing homes, rehab centers, and outpatient – be better coordinated and more transparent. At this point, that’s actually why we’ve built some of the technology that we have. Better connectivity would improve our efficiency. That’s probably top on our wish list.
DH: If we’re really truly as an industry going to do value-based care, we’ve got to have the ability to exchange that information in a secure and standard way. That capability really just does not exist in this industry vertical. There’s work coming out with the FHIR standard. We’re adopting the standards as they come out in the hopes that the industry will continue to move forward. Unfortunately, the industry moves fairly slowly, and so we’re having to develop a lot of workarounds. Getting data from providers and payers in a consistent and expected way is quite a challenge. As far as the portfolio maturing itself, we will continue to build purpose-based tools around our EHR to facilitate the care protocols and the care model that we have, because they don’t exist in the industry.