Health IT

CHIME Fall Forum Interview Series: Todd Rogow, CHCIO, Senior VP & CIO, Healthix

Todd M. Rogow, MPA, CHCIO

Todd M. Rogow, MPA, CHCIO Healthix

Healthix is the largest public health information exchange (HIE) in the nation, serving the most comprehensive range of organizations in New York, from the largest hospital systems to the smallest community health centers and physician practices. Healthix delivers data of more than 16 million patients to participant organizations that include hospitals and health systems, provider practices, behavioral health organizations, long-term and sub-acute care organizations, health plans, other public HIEs, and private HIEs. Todd Rogow, Senior VP & CIO, recently led the organization’s move from an outsourced resource model to an insourced technical team, including the implementation of a robust security program and SOC 2 Audit. In this interview, Todd elaborates on the benefits of building a mission-driven internal team to support the HIE, including improved scalability, nimbleness and responsiveness, but also cost effectiveness and innovation. Todd also shares his perspective on HIE funding models and sustainability, innovative approaches to patient identity and matching, leveraging predictive analytics to drive insight to the point of care, and the responsibility of the HIE in ensuring security and privacy.

CHIME Fall CIO Forum provides valuable education programming, tailored specifically to meet the needs of CIOs and other healthcare IT executives. Justin Campbell, of Galen Healthcare Solutions, had the opportunity to attend this year’s forum and interview CIOs from all over the country. Here is the next interview in the series:

Key Insights

When I joined Healthix two and a half years ago, I observed that we were losing ground because we were getting 11K new potential patient matches every day that required manual review.  With such a high volume, we couldn’t possibly keep up using a manual approach.

Having direct relationships with our vendors – whether they represent an application we leverage, hardware we run, or a service provider we work with – expedites the process of getting results by removing unnecessary overhead.

Insourcing the IT work has allowed us to become experts and facilitated a mission-driven, dedicated team that stays on top of our operations and growth. Being in this unique niche of healthcare IT and health information exchange really makes this approach advantageous.

Of key value to residents of New York is giving them access to their healthcare data. It’s something that we’ve taken steps to deliver through APIs made available to any of our participants that wish to tap into Healthix.  This enables them to make Healthix data available to patients through their own patient portals.

We believe that federal and state funding will continue to be a part of our sustainability model moving forward, although we can’t be sure of funding levels.  We are always exploring other revenue streams.

As a steward of PHI, Healthix understands that it is critically important to secure the data that we are entrusted to hold.  Technically, we do not own the data; it comes from a variety of participating organizations such as providers, payers, behavioral health, pharmacies, or in some cases Medicaid. It is therefore our obligation to protect it to the highest security standard we can offer.

Campbell: Tell me more about your bio, background, career trajectory, the organizations you’ve worked with, and the technologies they use.

Rogow: I’ve been involved in healthcare information technology for 15 years. I got my start working with electronic health record systems while as a contractor at Northrop Grumman, working for the Department of Defense. I helped to build their unique custom EHR, which was used by DoD, and spent several years enhancing that EHR product, from seeing its client-server application evolve, to helping create its first cloud hosted model. I then moved into the HIE space, spending over 5 years at HealthInfoNet, the statewide HIE for Maine. I was among the first five employees engaged there and saw the organization grow to a staff of 27. I led the redesign of the HIE from a technology perspective. One of the first things we did was to evaluate best-of-breed vendors to design an effective HIE solution for collecting data and providing real-time services to the participants, who are really the customer base.  The participants were comprised of clinicians in Maine’s healthcare community.

HealthInfoNet really shaped me and set me on a good path for what we’re doing here in New York. Going through that rebuilding experience and tackling scalability, having scaled the Maine HIE to be truly statewide, was impactful. In terms of the data we were collecting, the organizations we worked with ranged from behavioral health, with HIV sensitive data, to the common clinical data you would expect from reference labs or from hospitals or private practices.

When I joined Healthix, it was really to redesign the HIE, and begin a program to insource operations. For several years before I joined, the IT department was outsourced. One of the major tasks I was given was to build a team to handle the complexities of this business. We talk about Healthix as the largest public HIE in the nations. We really measure not just for the number of connections or data feeds we have built, but rather the size of the population we serve. At this point, we’re well over 16 million unique identities which contain clinical information. We have a lot of people who come in and out of New York City from all over the state, the country and even the world who may end up in our healthcare system.

On average there are 46 million messages coming into the Healthix system.  Over the last few years we’ve really focused on pushing data out. Like HealthInfoNet, Healthix is a real-time HIE, and that is where a lot of the value lies.  We have close to a half million real-time clinical alerts each month and push out over one hundred thousand continuity of care documents. In many cases, we build a tight integration into the EHR product, especially in those cases where the participants don’t have that capability, depending on the vendor they use.

Campbell: A well-rounded overview. I appreciate you reinforcing some of the high-level statistics you publish and highlighting some of the advanced work that’s occurring within the exchange today. If we could dive into one topic in particular, you mentioned managing more than 16MM lives. I want to touch on identity. You provided some detail around how a patient search is accomplished through demographics and MRN. Tell me a little bit about Healthix’s patient matching and identity management strategy, how exceptions might be handled, and what solutions you may leverage.

Rogow: I’ll provide you with another number. If you think of the variety of data sources that feed into Healthix – behavioral health, private practice, and hospitals – we get different medical record numbers from each of those organizations. As such, we have just over 58MM MRNs that we’ve brought in for the 7-8 years of data that we have. The challenge, as you pointed out, is really knowing that Todd Rogow is the same thing as T Rogow or just Todd Rogow who has gone to a different organization and has another unique identifier associated with him. We’ve been able to boil that down to close to 16MM unique identifiers and we have a couple of technologies in play that facilitate patient identity. We use IBM’s product, which was built by Initiate.

In addition, the velocity of matching associations wasn’t fast enough for us. We had a lot that fell into a gray area where we think they’re the same person, but they really need to be manually reviewed. As you can appreciate, this is extremely laborious. When I joined Healthix two and a half years ago, I observed that we were losing ground because we were getting 11K new potential patient matches every day that required manual review.  With such a high volume, we couldn’t possibly keep up using a manual approach. To automate the process, we contracted with Verato, a company that has a service that does something unique. They realized a while ago that there are a lot of public records for Todd Rogow. For example, I have an electricity bill, so there’s a public record of me and my address. There could be a credit agency that also has my name and my address and could include other things like a social security number, home phone number, or my date of birth. All of this is publicly available. They built an application that we reach out to as a service through an API, and we provide two identities for who we think may be the same person. We’re not certain, so we reach out to them and we ask them to query their public datasets from credit agencies, public utilities, etc., and come back with a recommendation on identity matching. Basically yes, maybe or no. It’s similar to what IBM is doing, but it’s another pass with more data that we don’t have access to.

With that, we have seen tremendous improvements. Not only have we dropped our manual approach of auditing these records individually, but we were able to go back and revisit our full backlog – anything which was a potential match. We were able to further collapse, by several million identities, and consolidate clinical records. From a clinician’s point of view, we’re now bringing extra clinical value around the proper identity of the patient and all of his/her records through that service. That’s been a really big improvement that we’ve made since I joined Healthix, and represents a new vendor that we’re working with very effectively.

Campbell: From sitting in on a New England HIMSS HIE advocacy panel event put on in early spring, outside security and privacy, identity is top of mind for HIEs. Thank you for elaborating on that. Shifting gears, you mentioned that you had out-sourced and then moved to an in-sourcing model. What challenges occurred with that, and what benefits did you realize as a result of moving to that model?

Rogow: I’d like to spend more time on the benefits, but let’s start with some of the challenges of moving from an outsourced to insourced resource model. A lot of companies go through the opposite – moving from insourced to an outsourced model. They think that outsourcing is better, only to swing the other way and insource. Just before I was hired, Healthix realized we needed more direct control over our destiny. What I mean is Healthix wanted to be very responsive to its customer base, and found that this was hard to do through 3rd party intermediaries. Having direct relationships with our vendors – whether they represent an application we leverage, hardware we run, or a service provider we work with – expedites the process of getting results by removing unnecessary overhead.

The other aspect is that Healthix didn’t feel that the growth we wanted to undertake could be accomplished without a change. Specifically, we didn’t feel that an outsourced vendor could keep pace with the scalability and amount of security required. Given the scope of the PHI stewardship responsibility of the organization, we felt that it was important to have that control.

As such, the organization engaged me to build a team and tasked me to insource our operations. Based on my prior experiences with HealthInfoNet in the state of Maine, I had familiarity with IBM Initiate for EMPI, and we leveraged Orion for the clinical data repository and clinical portal front end. However, Healthix implemented InterSystems HIE, so there was a little bit of a learning curve for me. The approach I took was to evaluate our system support needs and build a team that would address those needs. Certainly, there are some core roles you know you need to fill right out of the gate, but beyond that, there was examination of where resources were needed internally to be as responsive as possible to our customer base.

Once the core was in place, we directed focus on building new features and evaluating what resources we needed to tackle those initiatives. All-in-all, this approach offered tremendous benefits. We’ve witnessed enhanced scalability and quicker response times; a result of the direct relationship with our vendors. A major side benefit was the overall cost reduction we observed. We knew that if we were to try to scale to the same level where Healthix is today with an outsource arrangement, the costs would be too high. Insourcing the work allowed us to become experts and facilitated a mission-driven, dedicated team that stays on top of our operations and growth. Being in this unique niche of healthcare IT and health information exchange really makes this approach advantageous.

Campbell: I couldn’t agree with you more. It emphasizes the point that Ed Marx made at the NYSHIMSS meeting in that, it’s all about culture, and the ability to tap into that passion through a mission-driven team. The tough part is surely getting up to speed on a platform you aren’t necessarily familiar with, and identifying those roles you need to fill – whether its data governance and harmonization, or security and infrastructure. It’s great to hear that you were able to recognize some cost efficiencies as a result.

Rogow: Interestingly enough, we went through the same thing in Maine when I first joined HealthInfoNet. As I previously mentioned, when I first joined, I was among the first employees hired, and at that point, we had outsourced the IT portion as well. I brought that in-house. Being mission driven in this niche of healthcare really does attract the best people, and there is a lot of dedication that follows.  Ultimately, we are impacting patient care.

Campbell: I imagine you have countless stories of facilitating care coordination where key insights derived from the HIE are driven to the point of care as a result of the exchange.

Rogow: Absolutely. Both at HealthInfoNet and Healthix, I’ve heard stories where our staff goes out to visit with sites and they tell us how they are treating complex patients and how our service is being used to push insights out to them. It drives the point about mission home, and even though my staff are focused on keeping the system up and adding new capabilities and functionality to make it a more useful service, they believe, and I believe, that we are really saving lives. Not only saving lives, but also helping to improve healthcare for patients. That’s why we’re here and in this business. To have our staff get exposure to that is impactful when it comes to our mission.

Campbell: That covers the value proposition of HIEs quite well, but what about sustainability and solvency? Funding is an issue that comes up over and over again for both public and private HIEs. Whether it’s a subscription-based model that is used, or perhaps grants in play to prop up the HIE. What does sustainability look like for Healthix? You touched on having a series of clinical notifications planned, but how is that value funded?

Rogow: The state of New York is extremely supportive. The governor made a decision to support a statewide HIE through the use of federal funds that come in through CMS as well as state matching. Just before I joined, we entered into a period of 3 years where the state had allocated funding for us. 90% of our operational funds come from either federal or state funds. Before that, it was very similar to the model for Maine’s HealthInfoNet, where each participant, whether it be a hospital or private practice, would pay their portion of a service fee that would help fund the operation. We believe that federal and state funding will be a part of our sustainability model moving forward, although we can’t be sure of funding levels.  We are also exploring other revenue streams.

We’ve introduced new services, such as predictive analytics, focusing on the top 5-10% of the population that could be, or are the highest cost patients. We aim to get in front of the cost curve, and be proactively impactful, giving the care management teams of healthcare organizations an indications as to who the individuals are that we believe are likely to present in the ER or another inpatient setting, or have the potential of having a chronic condition. We highlight these patients so clinicians can effectively reach out to highest need patients. That is a Healthix service which customers are paying for today.

Other areas we’ve been exploring are customized real-time clinical event notifications. We offer a lot of the basic trigger events. For instance, if a patient presents in the ER, that will trigger an alert. However, if we’re able to provide a chief complaint, and other key data within that alert, that provides greater value to the provider or care manager.  Increasingly, we’re able to identify   services that our customers’ value and are willing to pay for. Healthix received funds for grants from various agencies, some at the New York City level, where we work on specific projects. As such, the grant money is really project-based and not a significant source of income.  Many feel that HIE shouldn’t rely on local city, state, or federal government picking up the entirety of the bill, but they do feel that there is a role for them to play in terms of funding. We’ve always thought of the three-legged stool in terms of funding – providers paying a service fee, the payers or insurance companies paying a fee, and the government contributing the remainder. The public services that we offer comprise the majority of our expenses, but we’re trying to make it a model where funding is more diversified across those we serve.

Campbell: It sounds like a very sound model. As you said, the point can’t be underscored enough in that it is truly a public service. What Healthix is doing is facilitating healthier New Yorkers, and healthier populations.

Rogow: We feel that there is a lot of untapped potential with delivery of services around predictive analytics and engaging providers or communities.  As they’re receiving a direct benefit from it, we can monetize those services. We certainly aim for a diversity of revenue stream, but having the backbone of government funding is critical. It’s helped us to mature as an organization, and to really show our value.

A critically important public value is giving the residents of New York access to their data. It’s something that we’ve taken multiple actions to deliver through APIs available to any of our participants that wish to tap into Healthix, so that they’re patient portals can make the data available for patients. Of course, all of this is contingent upon patients providing their consent, which is heavily controlled.

Campbell: That’s a great point. Being mission-driven, the most important entity in all of this is the patient. I myself use the MAHIway, and can appreciate the utility of having my chart available and being an active participant in my health. This is especially pronounced if you are managing many chronic conditions; having the HIE to lean on can be critical.

Rogow: It’s so important. We have a lot of HIEs across the country that are doing well overall and the service is getting more valuable as time goes by and technology improves. However, what’s really going to be a game-changer, is putting that control in the hands of the patient; when you’ve got a mobile device that provides you access to your records and allows for your records to be transportable. We are on the cusp of witnessing the patient really taking control of their records and leveraging that control to change healthcare. Not just the access to their clinical data, but providing awareness and contextual information around where to go for the best treatments, for instance.

Campbell: It certainly seems like there is a convergence taking place that will lead to data liberation. We know that the patients are clamoring for it given what’s at stake. Shifting gears a bit, at the NYSHIMSS meeting, the collaboration that occurs with other HIEs in New York through the SHIN-NY was highlighted. Can you touch on that and perhaps speak to other registries that you may integrate with?

Rogow: Starting at the highest level, we are connected to the Sequoia project, which is really the framework to exchange clinical information across the country. We’re also close to going live with the Veteran’s Administration for bi-directional exchange of clinical data with VA hospitals and ambulatory organizations. In terms of the SHIN-NY, it’s really a statewide clinical information exchange that’s comprised of each of the HIEs in the state. Right now, there are 8 of us serving the upstate and downstate regions. Since I’ve joined Healthix, we’ve witnessed a great deal of consolidation, which is a good thing. We know that people work, travel, reside and receive healthcare across geographic regions. Recognizing the size of the state’s population, and also the large geography, we have to collect data on behalf of the residents, regardless of the boundaries. This can be challenging, but there are services in place to identify that resident across the entirety of the state. This allows records to be exchanged within the state boundaries, and really offers a statewide support structure. Even though Healthix’s primary territory is New York City and Long Island, those boundaries go away when patients and providers retrieve data.

Going down to the next level, you mentioned attaching to different registries. We support several public registries. One in particular is the New York City Department of Health AIDS Institute. Their mission is to monitor the health and treatment of HIV+ individuals in our community and retain them in care. We identify HIV+ persons and the care they are receiving, so NYSDOH can focus public health surveillance efforts to ensure linkage to care, retention in care, antiretroviral therapy, and viral suppression.

Another registry we support that is kind of unique to New York State is what’s called eMOLST, around an end-of-life legal document. MOLST is a clinical process that emphasizes the discussion of patient’s goals for care, including shared medical decision-making between health care professionals and patients who are seriously ill or frail. The result is a standardized set of medical orders documented on the MOLST form that reflect the patient’s preferences for life-sustaining treatments. Our partnership with Excellus BCBS gives providers access, through the HIE, to the medical orders and wishes for an end of life patient.

We support a New York City Dept. of Health and Mental Hygiene service called NYCEPS – the New York Emergency Patient Search program. – After a mass casualty incident, a key concern is locating and assuring the safety of loved ones who may have been affected and cannot be easily reached or located. NYCEPs queries real-time patient data through Healthix, particularly information from encounters at acute care hospitals and nursing homes – all with the intention of facilitating family reunification. NYCEPS staff can help search for a missing person who may have been treated at a New York City hospital, thereby giving families’ one place to inquire about a family member. This potentially reduces the overwhelming number of phone calls to individual healthcare facilities which are already overburdened in times of crisis.

Campbell: Thank you for elaborating on those initiatives and advanced HIE use cases. It’s fascinating and compelling to see the many tentacles coming out of the HIE and the numerous entities that directly benefit. Any final thoughts you’d like to offer?

Rogow: I’m going to settle on security as the final thought, though it’s always at the very top of our minds. We will be undertaking the HITRUST certification by the end of next year. We’ve taken a lot of steps towards that third-party certification and have undertaken other measures, including going through a SOC 2 compliance and achieving that certification. These are critical when handling volumes of PHI.

As a steward of PHI, Healthix understands that it is critically important to secure the data that we are entrusted to hold.  Technically we do not own the data; it comes from a variety of participating organizations such as providers, payers, behavioral health, pharmacies, or in some cases Medicaid. It is however our obligation to protect it to the highest security standard we can offer.

With the recent press around ransomware attacks and digital security threats, Healthix takes this extremely seriously and dedicates a fair amount of resources and operating expense to implement the technology as well as secure the data. It’s an important message for people to understand: we consider this one of our highest priorities.

We talked about obtaining the patient’s consent in order for providers to access their record for treatment and quality improvement. The model we’ve implemented in New York State is consent to access (similar to an opt-in model).  This is what enables a clinician to look up a patient’s record. Typically, consent is obtained during the registration process. We have the proper technology in place so that we are able to control contextual access to the data, which could be sensitive, to ensure it is properly accessed by authorized and authenticated users.

Campbell: It’s reassuring to know that those safeguards are in place. It’s also refreshing to hear how serious Healthix takes their role as being a custodian of the data, especially in light of all of the other initiatives and operational functions. Considering the volume and velocity of the data, it must be top of mind, so it’s great to see that you folks are a leader in that area.

Rogow: When I came onboard, it was critical for me to bring on a Chief Information Security Officer. Our participant organizations and leadership continue to make this a priority.

Campbell: Thank you for taking time to speak with me. This has been extremely enlightening, and I am appreciative of you sharing your experiences, insight and wisdom.

About Todd Rogow

Todd M. Rogow, MPA, CHCIO joined Healthix in 2015 as the Senior Vice President and Chief Information Officer, where he is responsible for providing the vision, strategy and day-to-day operational leadership for all technical aspects of the company. Todd brings a wealth of knowledge and industry experience and has worked in the health information exchange space for over ten years.

Todd joined Healthix during a period of rapid growth and innovative change. He has fulfilled a critical role of building and leading Healthix’s Information Technology function, migrating its technology development and operations in-house through the implementation of next generation application software and completing the technical systems merger of several HIE organizations that now make up Healthix. He has driven a comprehensive security program at Healthix that includes the onboarding of a Chief Information Security Officer and achieving SOC2 security.

Before joining Healthix, Todd served as the Chief Technology Officer at HealthInfoNet, Maine’s State HIE. With 20+ years of experience, primarily in directing technical projects, he has provided business and consulting services to a range of Fortune 100 companies and many government agencies including the Department of Defense. Todd has served as a subject matter expert at conferences and on national and international panels and at conferences on the subject of healthcare technology. He has also authored a number of publications and case studies.

Todd has a Master’s in Public Administration and a BA from the University of Vermont. He is a certified Project Management Professional (PMP) and a 2012 graduate of the Hanley Center’s Health Leadership program. In 2016, he became one of only a few hundred Certified Healthcare CIOs in the nation.

About Justin Campbell

Justin is Vice President, Strategy, at Galen Healthcare Solutions. He is responsible for market intelligence, segmentation, business and market development and competitive strategy. Justin has been consulting in Health IT for over 10 years, guiding clients in the implementation, integration, and optimization of clinical systems. He has been on the front lines of system replacement and data migration is passionate about advancing interoperability in healthcare and harnessing analytical insights to realize improvements in patient care. Justin can be found on Twitter at @TJustinCampbell and LinkedIn.

CHIME Interview Series: David Parker, CIO/VP of IT, Magnolia Regional Health Center

David Parker, CIO

David Parker, CIO/VP of IT, HIM, & Ambulatory Informatics Magnolia Regional Health Center

Magnolia Regional Health Center, where David Parker is CIO/VP of Information Technology, is taking physician engagement to the next level. An annual physician retreat helps the health center gather information on what the providers need to perform the best care possible. In this interview, Parker shares how their providers drove the decision for a new EHR; how the MEDITECH platform has changed over the years; and the benefits of total provider engagement in a transition process. He also discusses the issues that arise with legacy systems and how archival is top of mind for the organization.

CHIME Fall CIO Forum provides valuable education programming, tailored specifically to meet the needs of CIOs and other healthcare IT executives. Justin Campbell, of Galen Healthcare Solutions, had the opportunity to attend this year’s forum and interview CIOs from all over the country. Here is the next interview in the series:

Key Insights

In our community, we enjoy our autonomy and do not want to get into the hip pocket of another big healthcare organization.

We followed the Ready program that MEDITECH offered to us and that significantly helped us achieve a successful implementation.

When we were running the MEDITECH Magic system, we had Magic on the acute side and GE Centricity on the ambulatory side. We had lab and radiology report interfaces, but aside from that, there was very little other integration between those systems.

Although physicians typically don’t want change and appreciating the monumental project that comes with replacing systems, they recognized this is the way for us to progress forward. 

That was the intent of the retreat – the findings from those breakout sessions. We took that and and determined how to start addressing this for the physicians. That’s what drove our selection process.

We know we can save our hospital money if we can pick an archival solution and sunset these legacy systems.

Campbell: Please tell me a little about Magnolia Regional Health Center?

Parker: Magnolia Regional Health Center is in the northeastern corner of Mississippi, about an hour and a half east of Memphis. We serve seven counties, with a population base of about 200,000 people. We’re licensed for 200 beds but operate 171 beds. We have roughly 200 physicians within our organization.

We have a graduate medical education program here, so we’re able to raise our own physicians, which has been very valuable for the past decade that the program has been in place. We’ve had one or more members from every graduating class either choose to continue their residency here or complete their residency program and then choose to set up shop with us; it’s been quite a blessing for our community. We are a city and county owned hospital, meaning we’re not associated with any other health organizations; we’re a standalone system. Our closest competitor is roughly an hour away. We’re fortunate we don’t have heavy competition in our neighborhood, although that’s starting to change. We’re starting to see a little more encroachment in our community from other healthcare systems. We’re all being pressured from different angles and trying to find ways to grow our systems; we must adjust and adapt.

Campbell: Great, thank you. What EMR system does MRHC currently leverage? With usability and productivity deficiencies currently driving replacement activity in the EMR market, do you have any plans to migrate platforms?

Parker: We’ve been a customer of MEDITECH since the early 90’s. Last year, we implemented MEDITECH’s 6.1 – their latest platform – on the acute side and this year, we’re implementing MEDITECH’s web ambulatory product; we’re a MEDITECH customer across the board. We have almost every single module that MEDITECH offers as it’s a good fit for a hospital of our size.

During the vendor selection process, several of our physicians wanted us to look at Epic as they had trained at much larger hospitals and knew the Epic platform and liked it. However, it was just not in the cards for us, as it was too expensive. Epic doesn’t sell to directly to community based hospitals like us, so the only option we had was to partner with another Epic hospital. We took that message back to our physician base. Here in our community we enjoy our autonomy and do not want to get into the hip pocket of another big healthcare organization, so we decided that was not an option.

The MEDITECH Magic platform has been a good product for us. We used it until it was getting long in the tooth. The younger doctors did not like the look and feel of Magic platform, so, we started visiting with MEDITECH and learning how they were moving forward. Their R&D dollars were not going towards Magic, but rather, they were going towards their new 6.x platform. As such, we selected that as our go-forward platform. We implemented that with MEDITECH’s Ready  methodology that they’ve put in place. It’s a project timeline and guide to follow with best practices for choosing your consultants; making sure those consultants know the system – they’re trained and certified; determining what teams to put together internally; how to backfill for them; and how to allow those teams to fully focus on the implementation. We followed the Ready program that MEDITECH offered to us and that significantly helped us achieve a successful implementation.

Campbell: It sounds like you made an outstanding decision to stick with the platform that you’ve long been on and served itself well to you. MEDITECH is well known for their stability and it’s great to hear that you’re advancing your use of their platform and adopting even more features through it. To that regard, tell me a little bit about how that decision was made in the context of interoperability demands? A lot of groups consolidate and move to a single solution so they don’t have to worry about interoperability within the enterprise, especially between care settings. Could you touch on your experience with that and how that’s handled within MRHC?

Parker: When we were running the MEDITECH Magic system, we had Magic on the acute side and then GE Centricity on the ambulatory side. We had lab and radiology report interfaces, but aside from that, there was very little other integration between those systems. As we moved forward, the doctors expressed the desire for one platform. When MEDITECH came on-site to do their demos, they also showed how this new product they were working on that would be fully integrated. The doctors who saw it could see the benefit of it. Although physicians typically don’t want change and appreciating the monumental project that comes with replacing systems, they recognized this is the way for us to progress forward.  Meaningful Use is here to stay and we decided we must continue plugging away at that and other anticipated regulatory measures. Our physicians recognized they’re going to require more and more use of electronic health records and having those systems integrated so we can harvest the data for reporting and analytics is critical to our success.

I’m very proud to say that our physicians drove us to make this move. We have an annual physician educational retreat, where we meet off-site and break into sessions. Out of all those different breakout sessions, there was the resounding sentiment from the 80 or so physicians who attended to replace MEDITECH Magic. It wasn’t that they pushed for a particular system, but they said, ‘We have used Magic for many years, we have made Meaningful Use Stage 1 and 2 with MEDITECH Magic, but we’re very frustrated with it. It’s time to go look for something else.’ That was the intent of the retreat, the findings from those breakout sessions. We took that and determined how to start addressing this for the physicians. That’s what drove our selection process for the next six months or so of 2014. We looked at the options including Epic, Cerner, McKesson, and MEDITECH. We made the doctors a part of that process and solicited their feedback. We also solicited input from all the other departments that it would affect.

Campbell: It’s truly profound that the providers drove the selection process, where you have engagement and they feel like they’re a part of it. Shifting gears a bit, can you tell me about data you migrated from GE Centricity? Did you abstract the data into the new system? Was there a data migration that took place? Is the GE Centricity system still running?

Parker: That is the one thing that’s been a little frustrating in this whole process, as MEDITECH does not have a migration path from Magic or any external system, so it was not an option to migrate data into the system. We still have GE Centricity running, as well as the Magic system, so we can still access historical data in those systems. MEDITECH 6.1 contains a link that allows you to contextually SSO to Magic, which is helpful, but we still need the icon for Centricity on the desktops for the users. Our plan over the next year is to start looking at how we are going to archive all of the data and retire the legacy systems. We have MEDITECH Magic data, we have MEDITECH Homecare Hospice product from years in the past, we have GE Centricity records, and we’ve got some other little systems that we need to archive. We need to be able to retire those legacy systems because right now we still maintain those servers and pay some licensing to keep the systems running.

Campbell: How does archival fit within the overall project of system upgrade and replacement?

Parker: We’ve been very focused on the 6.x implementation for the last two years. We kicked off at the beginning of 2015 with an implementation of the acute side, and that was roughly a 16-18-month project. Once we were live, we spent several months fixing things then shifted our focus to ambulatory. Now that ambulatory is live, we’ll probably spend a few months on enhancements and additional optimization opportunities. Then we’ll start looking at how to get rid of the technical debt that’s looming out there. We know we can save our hospital money if we can pick an archival solution and sunset these legacy systems.

Campbell: That sounds very logical. Shifting gears a bit, what is MRMC’s plans for population health management? Are you leveraging a solution today or do you have plans to? Or is it even something that’s applicable to your organization today?

Parker: It’s not too applicable right now. We do have the surveillance dashboards MEDITECH offers and we’re building them now, but don’t have them live yet. We’re evaluating incorporating those dashboards into the workflow, and we have an internal committee pursuing that initiative. Sepsis prevention is the big area that we’re focusing on right now. Once we get our arms around that, we’ll move onto other population health initiatives. We’re in discussions with a big hospital that’s about an hour south of us regarding collaboration through health information exchange. As we move that forward we’ll look at getting more population health data out of MEDITECH and into this bigger group of hospitals that’s forming a larger community.

Campbell: It’s how you survive in this value-based world. The data sharing must happen and that’s why data blocking is such a huge topic. The patients are demanding that the data follow them, but the infrastructure may not be in place to allow it to happen. Do you have a comment on any other projects that might be ongoing at the organization once you’ve completed the implementation?

Parker: One of the next big large initiatives we plan on tackling is clinical documentation improvement. We recently purchased Nuance’s Clintegrity product and we’ll soon be focusing on getting that up and going. We think that’s a game changer for us and our physicians have been clamoring for something like this. We survived the switch over from ICD-10, but there’s so much more that we could be doing to improve documentation, to code our charts better and to accurately reflect the health of our patients. We were very disappointed in our health grade score, which surprised us, but as we started digging into the data it was clear to us that we are not doing a good job of documenting just how sick our patients are. It looks like they’re not very sick, and they come in and get much sicker, or pass away, and we haven’t done a good job to document that these patients were very sick when they presented at the hospital to begin with. The CDI program that we’re putting into place with Nuance will take us roughly six months to get it in place, but we think it will be a positive change for us.

About David Parker
David Parker serves as CIO/VP of IT for Magnolia Regional Medical Center, a non-profit, city owned, HIMSS EMRAM Stage 6, 200-bed acute care hospital located in Corinth, MS. Mr. Parker leads an IT team of 24 employees and is responsible for IT management, project leadership, budgeting, & strategic planning. Mr. Parker currently oversees upgrading of platforms and operating systems as a part of MRHC’s initiative to become a HIMSS EMRAM Stage 7 hospital.

Prior to his current position, Mr. Parker served as an IT director for a smaller health care system in Oklahoma for 10 years. Mr. Parker has also held positions where responsibilities included support of a local hospital finance system and electronic medical records implementation at a health care facility.

Mr. Parker holds a BS in Finance from Texas A&M University. 

About Justin Campbell
Justin is Vice President, Strategy, at Galen Healthcare Solutions. He is responsible for market intelligence, segmentation, business and market development and competitive strategy. Justin has been consulting in Health IT for over 10 years, guiding clients in the implementation, integration, and optimization of clinical systems. He has been on the front lines of system replacement and data migration is passionate about advancing interoperability in healthcare and harnessing analytical insights to realize improvements in patient care. Justin can be found on Twitter at @TJustinCampbell and LinkedIn.

CHIME Fall Forum Interview Series: Susan Carman, CIO, United Health Services

susan-morreale-cio

Susan Carman, CIO, United Health Services

Susan Carman is no stranger to the fast-paced change of the healthcare IT world. United Health Services, where Carman is currently CIO, is in the process of overhauling their IT strategic plan, including a potential replacement of their best of breed EMR/EHR for an enterprise solution. Since the current inpatient EMR was only implemented 3 years ago, Carman is faced with the tough decision of yet another system replacement, both from a fiscal and employee fatigue point-of-view. She discusses the implications behind their review of the enterprise; why being best of breed and having a data warehouse doesn’t always jive; and her key takeaways from CHIME.  She also provides sage advice to those startups and digital health organizations trying to get their solutions in front of healthcare IT decision makers.

CHIME Fall CIO Forum provides valuable education programming, tailored specifically to meet the needs of CIOs and other healthcare IT executives. Justin Campbell, of Galen Healthcare Solutions, had the opportunity to attend this year’s forum and interview CIOs from all over the country. Here is the next interview in the series:

Key Insights

When I first came in about a year ago, there was no true inventory of applications. We used a number of avenues to figure out exactly how many we actually had.

We’re a best of breed shop, the kind you don’t see much of any more. It’s causing us a lot of issues because we have about 200 interfaces running information back and forth. It’s very resource intensive and expensive to run.

We’re trying to figure out if we can connect directly to our data warehouse as opposed to connecting to the source systems directly. That’s a big question mark for us right now.

With this information the IT governance team decided that the best approach was to develop a 3 to 5-year Strategic Plan.  The most important component of that plan was to decide on a future EMR roadmap.  Our biggest challenge is meeting our integration goals but doing it in a fiscally responsible manner.

Sitting around the table with other CIOs—gathering the information about where they’re at with their application portfolio, best practices, and things of that sort—was immensely helpful to me. I think it’s mutually beneficial because clearly the vendors are doing this to get some information for their business.

Campbell: Can you give me a little background about you and your organization?

Carman: We have about 380 physicians, 4 hospitals – 2 of which have over 200 beds – 2 smaller hospitals, and 62 clinics.  One of the biggest things we’re working on is our IT strategic plan, so a lot of the questions I’ll be answering today will, unfortunately, be changing over the next year or two.

Campbell: No problem, that’s to be expected. Along the lines of the strategic plan, how many applications does the organization have within the portfolio and how do you manage those today? Do you have an enterprise portfolio management system or is it managed by spreadsheet?

Carman: That’s a great question and going back to talking about being in a state of transition, this is something we’re working on, and spending a lot of time on right now. We have approximately 175 applications that we run. When I first came in about a year ago, there was no true inventory of those applications. We used a number of avenues to figure out exactly how many we actually had. It was sort of an unknown, but I certainly knew there was quite a few of them.

We’re a best of breed shop, the kind you don’t see much of any more. It’s causing us a lot of issues because we have about 200 interfaces running information back and forth. It’s very resource intensive and expensive to run. The application portfolio is basically being kept on a spreadsheet.

Another initiative we’re going to be starting next year is we’re purchasing an ITSM system. A lot of our application portfolio should be kept in that system once it’s implemented rather than keeping it on a spreadsheet. We’re bringing in a consulting company for the first half of next year to help us work on adopting ITIL best practices. The implementation of the ITSM system will be about midway 2017. We’re hoping for some great things as far as doing a better job of keeping track of our applications, providing better customer service, and improved change management.  The last step will be application rationalization to sunset and retire anything that we no longer need.

Campbell: Very good, and speaking of application rationalization, do you leverage an archival solution today? Do you have a data warehouse that it’s shipped off to? Or are you simply taking a copy of the database and storing it on your servers? Could you elaborate on that topic?

Carman:  The business intelligence and analytics software is overseen by the CMIO at UHS.  It is not part of the Information Services department.  Our CMIO is not only a practicing cardiologist, he is an expert in the development of applications and business intelligence. A lot of what we have at UHS is homegrown.  We have a separate Business Intelligence department. We utilize a lot of students from Binghamton University that are working toward their Master’s degree or PhD and need a real life project. We do have a data warehouse.  We are striving to get to a point where our data warehouse is our one true source of information.  We are currently working on our data accuracy via a new data governance model.

Campbell: Right, a lot of organizations are dealing with that, those point solutions, especially if you’re best of breed. How do you integrate that ETL process with the data warehouse?

Carman: We just purchased Watson Health as a population health solution, and we’re going to be implementing that next year as well.  We’re trying to figure out if we can connect directly to our data warehouse as opposed to connecting to the source systems directly. That’s a big question mark for us right now. This would require ensuring all the data is accurate before it gets into the data warehouse and that all of the data is being sent and in a timely fashion.  We don’t have all the data normalized and some things don’t go to the data warehouse at all for various reasons. Sometimes it’s limitations of the applications, but there’s a variety of reasons why not everything is going there right now. We would like to avoid connecting to every source database but we will likely have no other choice.

Campbell: I can appreciate that. You want that data warehouse to serve as the hub, but oftentimes there may be data that’s missing in the ETL process from the source system. I know you spoke of potential replacement of the EMR/EHR. How long has the organization been on the current EMR on the inpatient side and the EMR on the outpatient side?

Carman: We’ve been on NextGen for 12 years and it has outlived its usefulness. It is no longer supporting our physicians in an efficient manner. It’s come to a point where the writing is on the wall with NextGen. We need an ambulatory information system that is more adaptable to the differences in our specialty practices.

As far as the inpatient side, it’s only been 3 years since Soarian was implemented. Now of course we got the bad news that Cerner is only going to support the Soarian Clinicals for 5 more years and we’re 2 years into that.  When I first came on-board, one of the things I was charged with was to survey the ambulatory information system and gather information from the physicians as to whether they wanted to enhance the current system or replace it. Overwhelmingly they wanted to replace the current system.  Since we had just gotten word from Cerner that Soarian Clinicals would only be supported for a maximum of 5 more years, it made sense to review our entire enterprise. With this information the IT governance team decided that the best approach was to develop a 3 to 5 year Strategic Plan.  The most important component of that plan was to decide on a future EMR roadmap.  Our biggest challenge is meeting our integration goals but doing it in a fiscally responsible manner.

Campbell: Right and, to share with you, that’s the sentiment I heard from several folks while at CHIME. There’s just a lot of ‘where’s the money coming from?’ ‘Sure the organization wants to invest in a new application or maybe replace the EMR, but show me how we’re going to be solvent in whole once we do that.’

Carman:   Our Senior Leadership team mostly agreed that we need a fully integrated solution.  Our first step in that process is an EMR Financial Feasibility study.  We need to see exactly what we are up against when it comes to total cost of ownership.

Campbell: Shifting gears to CHIME, tell me about your experience. What did you look get out of it? What were some of the things you appreciate about the event?

Carman: I loved it. I can say overall, I thought it was fantastic. I got more out of the focus groups than anything else. Sitting around the table with other CIOs—gathering the information about where they’re at with their application portfolio, best practices, and things of that sort—was immensely helpful to me. I think it’s mutually beneficial because clearly the vendors are doing this to get some information for their business. Out of the entire week, that was where I felt I got the most information.

I certainly had a lot of questions about converting to an enterprise wide solution from best of breed, I spoke to a lot of different people about their situation. I even spoke to the #3 person at Cerner about customers who have transitioned from Soarian to Cerner.   I left there with a good feeling that we were performing the correct analysis. It was a great validation for me, and a very worthwhile trip.

Campbell: That’s excellent to hear. Aside from the networking aspect, like you said, you’re able to test hypotheses and synthesize strategies. Perhaps you could offer some insight to our audience at Health IT & mHealth, which is primarily focused on startups and digital health companies – a market that is overwhelmed by patient engagement solutions. For your organization that’s traditionally been best of breed, and maybe looking to an enterprise wide solution, tell me a little bit about how you would ever consider a digital health startup solution? What would be your criteria for evaluating that?

Carman: Target what the bigger companies don’t have or don’t do well. I think what you’re going to see is, if you’ve got a Cerner or an Epic Enterprise solution and they’re selling that same product, you’re not going to win because CIOs will gravitate toward their core vendor. They want to just go to one person; they don’t want to have numerous vendors that they have to deal with on a daily basis because it takes up so much time for a CIO. If there’s some sort of niche that you can find, where the vendor has either done a lousy job of providing the solution or they don’t do it at all – and certainly Epic has several things that they don’t do – concentrate on those areas. When you’re coming to the table you can say ‘look, you don’t have this currently, you’re not able to get it from your core vendor, and we have a great solution.’ I think that’s how you get CIOs to listen to you.

Another piece of advice I would give is always come to the table with how you’re going to make that CIOs life easier. First and foremost, what are you going to do to solve the problem that organization is experiencing?  Investigate who is in need of the solution before you go and invest your time.  Make sure to target your audience in that way. I get dozens and dozens of emails a day about things that aren’t relevant to me.  I think, ‘goodness these people are wasting a lot of time, I’m not responding and so on and so forth’. I think a lot of the smaller vendors need to do a better job of targeting their audience. Every now and then I get an email and I say ‘hey you know what I am going to respond. I don’t normally respond, but this person really got my attention, it’s something I need, they clearly understood my business here and what we do.’ It wasn’t a generic email.

Campbell: That’s a great point and great advice for anyone in the space. Any other closing thoughts you wanted to offer around the CHIME event or any organizational newsworthy items?

Carman: At this point, one of the big things we are grappling with, is what do we do with MACRA and MIPS and how do we transition. There’s no more PQRS, and MU is fading away. It’s a big question mark for us, and now with the new administration coming into play I think we’re all kind of scratching our heads saying ‘what does that mean for healthcare IT?’ With Obamacare up in the air, I’m sure a lot of CIOs are sitting back and saying ‘does that effect MACRA, MIPS, and everything else coming down the pipeline? What’s going to change?’

CHIME Fall Forum Interview Series: David J. Runt, CIO, Contra Costa Health Services

david-runt-cio

David Runt; CIO, FHIMSS, CPHIMS, Contra Costa Health Services

One Patient. One Record. That’s the philosophy at Contra Costa Health Services, where David Runt is Chief Information Officer. Their uniquely structured health department has everything health-related under a singular patient record, all fueled by Epic. They’re most healthcare organization’s dream. Runt considers himself fortunate to be a part of it. In this interview, he discusses CCHS’s transition from build to buy, including the sunset of a large amount of legacy systems; their data warehouse good fortune; why their data retention standard is “forever”; and how CHIME stands out from other industry events.

CHIME Fall CIO Forum provides valuable education programming, tailored specifically to meet the needs of CIOs and other healthcare IT executives. Justin Campbell, of Galen Healthcare Solutions, had the opportunity to attend this year’s forum and interview CIOs from all over the country. Here is the next interview in the series:

Key Insights

Our philosophy is one patient, one record.

Use those talented internal development resources to build niche product for customers that you can’t buy.

However, that served as the foundation to feed data into Epic. All of our data is going into the warehouse. It’s fantastic, the amount of data we have at our fingertips now. It’s revolutionizing our capabilities.

Our philosophy is to retain the data forever. Given our model, we don’t lose patients, so to speak, to our competitors.

Don’t be everything for everybody. Pick your niche. Get successful there and make a name for yourself there.

Campbell: Please tell me a bit of about yourself and your organization.

Runt: I’ve been with Contra Costa Health Services for eight years. I was in Arizona for ten years prior to that. Contra Costa Health Services is a county health department.  We are probably the only one in California that’s structured the way we are, in that everything “health” related falls under one patient record. This means that, public health, mental health, EMS, hazmat, health plan, hospital clinics, everything falls under one health officer. We provide health in five jails across the county as well. In our organization, public health reporting falls under one umbrella, with health services & traditional hospital clinic structure falling under the other umbrella. As a result, we have a lot of flexibility in managing health across the enterprise.

Our philosophy is one patient, one record. Whether that patient is seen in our hospital, our clinic, in our psych emergency room, in the jail, in public health – we know everything about them. We’re fortunate that we have that data in one location and we don’t have to coerce the data out of a perceived competitor such as the health plan or public health, like the other county health IT departments have to do.

Campbell: What platform do you leverage for the EHR?

Runt: We’re leveraging Epic across the enterprise. We’ve got it in the five detention facilities, as well as in our health plan. We’ve deployed it in behavioral health, in our own behavioral health clinics, and made it available to our network providers as well. We’re not selling services, but rather we’re giving the providers portal access for their patients. In this next year or so, we are rolling it out more into the public health space.

Campbell: That’s exactly where your peers want to be. In talking to a lot of other folks, they have one system on the outpatient side, and a different system on the inpatient side. As a result, harmonization of nomenclatures and dictionaries between the two systems, as well as reconciliation of the data poses to be an ongoing challenge.

Runt: Not to mention your health plans on another system! I could not imagine trying to get payer information on your own patients with that model. My counterparts are just pulling their hair out and feel fortunate to be where we are.

Campbell: It allows you to focus for sure. Speaking of that focus, tell me a little bit about your application portfolio, outside of EHR. How big is the portfolio? Are there legacy systems that are sort of sitting in the corner?

Runt: Everybody has the legacy systems. When I got there, we were a cat and dog shop. We were a development shop and I had 50 developers on staff. They were actually writing—which got killed before I got there—a patient appointment system. Really that’s a commodity, you buy that, so a lot of these things were sunset when I got there. We turned the organization into a buy instead of a build. We had a LOT of cats and dogs, best of breed systems. We were the poster child for the best of breed shop. As a result, bringing in Epic proved to be tremendous, as we sunset somewhere around 75-100 systems. Some of these were just little Access databases sitting on someone’s PC, but we were able to sunset a lot of that and reduce a lot of the ongoing licensing, maintenance and support costs.

We’ve gotten out of the “build your own stuff” mentality except for specialty areas such as environmental health, where you have an underground storage tank inspection system for instance. You can’t buy applications like that, so we build those. We use those talented internal development resources to build niche products for our customers that we can’t buy.

I am very fortunate that we’ve had a data warehouse for well over 15, maybe 20, years. It wasn’t very populated, in fact it was primarily financial data, because of these cat and dog systems. However, that served as the foundation to feed data into Epic. All of our data is going into the warehouse. It’s fantastic, the amount of data we have at our fingertips now. It’s revolutionizing our capabilities.

Campbell: Data liquidity cures all. Now when you went through that consolidation process, did you archive the data into that data warehouse?

Runt: Most of the data was already there. Those cat and dog systems, the ones that were sitting on somebody’s desk, we probably didn’t need that data anyway. It was Excel spreadsheets, or Access databases – those really didn’t have an impact on the business overall. Some of those systems that we sunset, we just took the data, imported it into SQL and populate it into the warehouse.

Campbell: That’s a critical point. The archival strategy whitepaper we discussed examines the tradeoffs, when you extract data from a legacy system and you store it because inherently you’re changing the shape of that data. You’re changing the shape of it to store that data; you’re changing the shape of it to present that data. There’s also metadata and audit trails considerations.  All that said, looking back on what was your archival strategy and how did you handle it?

Runt: We did not have an EHR. We had MEDITECH primarily for billing and ADT functions, and consequently, we didn’t have much clinical data in there. A year after I got there, CCHS finished a painful implementation of ED, and that’s the only clinical data we had. That was already in the data warehouse – we took it from MEDITECH and populated the warehouse. Other than that, it was billing information from our third party billing systems. We really didn’t change the shape of the data that much.

Campbell: Shifting gears a little bit, I’d like to get your perspective on data retention requirements. I’ve been talking to some of your peers about the fact that most states require persistence of the data for 7-10 years. A lot of EHR vendors don’t allow the ability to purge and there may be some instances when you do you want to purge that data. For instance, when that patient is no longer yours, you’re no longer seeing that patient, yet you still have their data so there’s still some liability associated with that if there’s an eDiscovery request.

Runt: Our philosophy is to retain the data forever. Given our model, we don’t lose patients, so to speak, to our competitors. Being the county, we serve a unique, well-defined, patient population. Yes, some people will flow in and out of the system, depending on their economic or social background, but for the most part we retain our patients. So we want that birth to death record to be available, wherever that patient may present. The patient could move around the county, but they’re still going to be seen in one of our clinics.

Campbell: And does Epic natively allow for ease of eDiscovery if you have an inquiry for a record? Or do you go to the data warehouse?

Runt: We go to the data warehouse for that.

Campbell: Shifting gears again, let’s talk a little bit about CHIME. Certainly the networking aspect alone is unparalleled. Tell me a little bit about how long you’ve been coming to CHIME and what you are looking to get out of it, specifically, this year as it relates to your initiatives.

Runt: I’ve been in healthcare IT for 36 years and have been a CHIME member for about 15 years or so, maybe longer. It’s much better than other industry events in that you have the one-on-one interaction with your peers. CHIME gives you the opportunity to reconnect with peers whom you have a relationship, and also connect with the important vendors. Other events have 20,000 vendors there and half of them won’t be around the next year; half of them aren’t relevant to the types of things I’m focused on.

It seems that only the “cream of the vendors” are associated with CHIME. It gives you the opportunity to interact with them, have meetings and meals with them, social interaction with them, which for me has been a benefit. Unlike CHIME, I don’t go to other events for the sessions, I go there to talk to vendors. The sessions at CHIME are different from a lot of industry events and conferences in that they are educational and not sales driven.

Campbell: It’s just more focused here.

Runt: Exactly. It is more focused, and I wish there were more organizations like CHIME for non-healthcare because there are a lot of good ideas out there in the marketplace around data, around PMO activities, and other things that are not purely healthcare-related.

Campbell: Lastly, focusing on HealthIT & mHealth’s audience, which are startups and entrepreneurs, do you have pieces of advice for them? How does a niche, small, little-known startup- vendor capture your attention? A lot of these startups are jumping into the fray with patient engagement solutions and if you really look at it, are they really solving a problem for their end customers? This is a very broad question, but if you have any piece of advice for startups and entrepreneurs entering into the healthcare space what would it be?

Runt: Don’t be everything for everybody. Pick your niche. Get successful there and make a name for yourself. We’re doing business with some small vendors who were probably startups five years ago. They’re small, they’ve got their niche product or their niche service, and they don’t want to be everything to everybody. Stay true to your initial idea and your initial concept. Answer the questions of ‘Why did you go into business? What need does your services or solution address?’

This interview has been edited and condensed.

CHIME Fall Forum Interview Series: Perry Horner, CIO, Adelante Healthcare

Perry Horner; CIO, Adelante Healthcare

Perry Horner; CIO, Adelante Healthcare

As a CIO of major healthcare system, one of the most important lessons Perry Horner learned was when he found himself personally forced into the complicated world of connected health. With his new motto of ‘keep it simple,’ Horner is taking on the connected health world full force, implementing new and affordable technologies, always looking three years ahead. In this interview, Horner talks about Adelante’s EHR transition from Vista to Nextgen; his data governance strategy; and what he’s looking for in a practice management system solution. He also discusses his compelling personal story with the healthcare system and the improvements he’s pushing for to make solutions simpler and more affordable to all.

CHIME Fall CIO Forum provides valuable education programming, tailored specifically to meet the needs of CIOs and other healthcare IT executives. Justin Campbell, of Galen Healthcare Solutions, had the opportunity to attend this year’s forum and interview CIOs from all over the country. Here is the first interview in the series:

Key Insights

I never used to be an agile project management style person, but over time I’ve changed and I see the same thing with software. Gone are the large monolithic products, where they do everything. Stick with your core competencies. If you’re doing something well, partner with someone who’s doing the other part well. Make one conglomerate product. That’s where I’m hoping to see things go.

I was in the ICU for 10 days, and when I got out I had 5 months to recover before I could go back to work. During those 5 months I went through the ringer as far as what a patient experiences, and I was taking notes. Mental notes and physical notes as to what I was seeing, how things were being treated, keeping a focus on technology. When I got back, I wrote a 5-year plan for Adelante and I infused into it what is now called “Connected Health”.

I’m tired of hearing “I have an app for that.” I’m experiencing app overload. I’m kind of reversing thoughts, back to making it simple.

Forget about trying to capture the whole supply chain. Find your place on the supply chain and just focus on it. Do the best you can there.

Campbell: Let’s start with a little background. I saw that you actually come from the vendor side, that’s probably a helpful perspective to have when working for a provider organization. Tell me a little about yourself and where you’re working at now.

Horner: In my previous life I spent 17 years at ASU, I was the Head of the Library Systems, Technology Support and Development department. I jumped into healthcare with Adelante Healthcare because they intrigued me with a position opening, a Linux network administrator—which there is no such thing, so I was curious about what they were doing. The CEO immediately gave me an offer to head the entire technology operation there so I took that.

I inherited their EHR at the time. At that time Adelante had 7 locations. The first site they brought up using the open sourced version of the VA Vista EHR, and with the success of launching that site, we received a HRSA HIT Rapid Grant. Two Grants actually: one to form a HCCN, and the other to implement an EHR for all those members. Two other community health centers joined us and we became an HCCN and implemented Vista for all of them including Adelante, which was a member of HCCN. I was then hired by the network, which was called the Community Health Open Source Network, as their CIO. I hired three others and we developed out our support department for all the members. That lasted 18 months and then our largest contributing member pulled out and it was not sustainable anymore, so we decommissioned everything.

Adelante Healthcare was on Vista and I knew we could do better, even though as a technologist I was in love with the Vista system. It’s a MUMPs database, which is basically a flat file database, beautifully written, I’m amazed at it. As far as the end product and the application, for us, there was a little of the square peg round hole that had taken place. When Meaningful Use, the whole Accountable Care Act, etc, came about, Vista kind of squeaked by. My crystal ball said that this is going to give us problems all the way down the road with stage 1,2,3, and having to actually develop all the little kludges around it, because there are no companies that are doing it. So I started a sunset on Vista and we RFP’ed for a new EHR.  We settled on NextGen as our EHR.

Before my time, when they were working on implementing the first site with Vista, it was done in a garage development style. The EHR that Adelante was using included traces of experimentation from the beginning and there wasn’t a demo or learning system created. So both the CMO and myself were not comfortable with bringing over the data into NextGen. Instead, we transferred over about 12 key demographics for each patient over into NextGen and left the entire chart in Vista. For the next year, providers would go back to Vista when they saw a patient that had a NextGen record and extract certain elements out of their chart and put them into NextGen. The whole patient cycle included about 38,000 unique patients.

When that was done we started on the sunset project for Vista. It was too expensive and cumbersome to maintain the operational server. So I had to get all that data out of there and into a system diagnostics machine-readable format. I wanted everything to be indexed using four different identifiers, so you could do reverse look-ups using: date of birth, last name, social security number, and the medical record number. I wanted the format to be in discrete XML – specially in a CCD format. Every single patient record needed its own directory with sub-directories of all the imaging data and any other files. We needed two formats: the XML and then a presentation mode – everything in a PDF. In order to accomplish this, I contracted with a programmer, who over two years, finalized the extraction. Vista is so complex with its filing system hierarchy within that database. The referencing and finding the linkage of every piece of that patient’s record was challenging, as it’s scattered everywhere.  I was quite impressed that we were 100% able to extract everything, reformat it and make it useable.

That was our EHR at that time. At that time too, our practice management system was GE Centricity Practice Solutions, so there’s my next project next year, getting that data out. Luckily that’s a SQL database, much easier than doing it with MUMPS.

Campbell: Tell me a little about data governance in your organization and how that is handled

Horner: We’ve got a plan for 2017: creating an information governance committee. In the meantime, everything is more specialized—a privacy committee, security committee, the various others around compliance, quality assurance—we want to bring all of that up into the next level of the leadership decision-making body and have an overarching Information Governance Technology Steering Committee.

Campbell: Do you have challenges at all with interdepartmental nomenclature, mapping? How is that handled?

Horner: Very wild-wild west style. We are small enough at the moment where we can still talk to each other and accommodate. Ten years from now we won’t be. We definitely need to have common definitions, common protocols. That’s why in 2017 the information governance will form and set the framework to start development from there. I’m pretty pleased that AHIMA has really taken the lead in producing a lot of resources around information and data governance.

Campbell: And speaking of the portfolio, how many applications do you have in it today? Do you leverage any portfolio management applications or is it small enough where it can be optimized?

Horner: Small enough at the moment.  Right now, except for our HR system and a credentialing application, everything is on premise. In two years, we will have outgrown our corporate headquarters, so we’re building a new facility and this time around we’re not building a data center in it. The project kicks-off in December, and we are slowly migrating hosting out to third party, starting with the low hanging fruit – file services, those type of things – with the EHR at the end.

Campbell: Mission critical applications last.

Horner: Yes, we designed our Mesa Facility as our fail over location, so we do have redundancy there. We’ll continue to use that for fail over but for our primary, we want it in an external data center. And if anything is platform-as-a-service, software-as-a-service, we are always entertaining that. Anything that can accommodate a heterogeneous environment.

Anything that’s HTML5, a no SQL database back-end, that true software-as-a-service environment, that’s what I keep my eye on. Any vendor that’s moving that route, has a product there, is large enough to actually interface with others. I never used to be an agile project management style person, but over time I’ve changed and I see the same thing with software. Gone are the large monolithic products, where they do everything. Stick with your core competencies. If you’re doing something well, partner with someone who’s doing the other part well. Make one conglomerate product. That’s where I’m hoping to see things go.

Campbell: I did some research and saw that there was an initiative, back in May, where you went out for bid with some other organizations for application development

Horner: Yes, that is on the connected health end. That was the Phoenix Chamber of Commerce’s Reverse Pitch for Healthcare. Speaking of connected health, in the beginning of 2014, I went in to see my doctor, who happens to be an Adelante Healthcare provider. My heart was one minute going 160 beats, next minute going 70, just off and on. It was a nuisance to me and it had been doing that for about 4 weeks, during December, during the holidays. She hooked me up to some diagnostics and after about 20 minutes of checking readings, she immediately looked at my wife and said, ‘if he drove, he’s not, you’re driving him straight to the hospital; he’s to be admitted to the ED.’

For the next 48 hours’ things were looking bleak. I was on my last leg. I was feeling fine until that Monday evening when I was in the hospital. My heart was failing – down to 15% capacity, and my family was told I had maybe 48 hours, and to start making preparations. I had a team of twelve different specialists all trying to figure out what was going on, and finally it was a cardiologist who figured it out. It was my adrenal gland. One of them was causing this and he came up with a cocktail that actually worked, stabilized my heart rate and threw me on the road to recovery.

I was in the ICU for 10 days, and when I got out I had 5 months to recover before I could go back to work. During those 5 months I went through the ringer as far as what a patient experiences, and I was taking notes. Mental notes and physical notes as to what I was seeing, how things were being treated, keeping a focus on technology. When I got back, I wrote a 5-year plan for Adelante and I infused into it what is now called “Connected Health”.

I was given a great care plan, things that I had to do at home, and I didn’t have an at home nurse, so that responsibility was given to my wife. She would make sure I was taking my blood pressure every 4 hours, and taking these pills in the morning, these pills in the afternoon, I was walking around for x amount of time, and ensuring I was eating these low salt foods. Two weeks later she was tired of it, she was angry with me, and I was tired of it too. I was then slacking back into a mode where I wasn’t doing everything and my specialists and my PCP weren’t seeing me every other day to monitor. I was on a schedule to come back and they would evaluate me and see what my progress was. I was like “there’s got to be something to keep me motivated and also keep that care team informed early, in case something goes wrong.”

A lot of times providers experiment. You’re on one medication and then they’ll check-in maybe three months later to see if that worked, and if not, it is changed. Now take this medication. Well it didn’t work in the first 72 hours, it’s not going to work for the remaining 90 days. Connected health solves this by having those monitoring and diagnostic devices in the home. It’s nothing new for the hospital world, in fact it’s been around for a while. You get discharged and you take a package home, that’s got a blood pressure cuff, a thermometer, a scale, or whatever, all connected via an app to your phone, to the internet. For me though, knowing how tired and wiped I was from the hospital experience, I cannot see how somebody who wasn’t a CIO would go through the process of connecting all of these devices, having to always spark up a tablet or a phone in order to see readings and do things to communicate. I wanted it to be as simple as possible. For example, with my scale in the bathroom, I could replace it with a connected device whereby, I just set this scale down, I don’t have to program it, do anything to it, I just have to step on it. Everything should be done already. The reading should be able to go back to my provider seamlessly. Same thing with the blood pressure cuff, I just want to press the button and that’s it. Nothing else.

I went to HIMSS Connected Health Conferences, which was really a bunch of developers trying to figure out what’s going to play into this world, with not very many established products yet. I desired a product that was just dirt-simple. For our population, 50% of our patients are on state Medicaid, which means they’re in financial distress. I want our devices to be as [affordable] as possible. It just needs to be what it is. That was my ask. Create this…and there was a company that actually did. Carematix collaborated with Verizon to develop a solution. So now you’ve got your carrier service, and you’ve got your product. The transmission doesn’t go directly to the cloud from the scale. The scale actually communicates to this little hub device that is plugged into the wall or any outlet, hidden away. The patient plugs into the power, steps on the scale, anywhere in the household. They don’t need to program anything, don’t need to worry about it, don’t ever have to remember that they have the little box plugged into the outlet ever again. Everything works and it’s affordable. That was the other piece. Carematix has priced their product at a dollar – a dollar a day per patient. That’s $30 a month for two devices. The cloud based management system, the whole shebang, no carrier fees, nothing – that is everything. So for us, we’re looking at $360 per patient per year. That is sweet. Problem solved.

But I had to wait 2 years for something like that to appear, and none of this is new technology. We’re going to do a pilot for two years now too. We have a control group – same chronic conditions, same age, everything very similar to its pair experimental group. The experimental group gets the devices while tthe control group receives traditional care. Over time we’re going to see if we can improve that patient’s health by early warning, being able to follow up immediately when we see something, and being able to contact that patient to see if we can change something. Improve that health much faster than the traditional way. If so, money well spent. To circle back around to your original question, that’s what that initiative back in May was about.

Campbell: It’s seems you’re well in tune to startups. What learnings did you have with interacting with some of those startups and what advice would you have for folks in those areas that would give them a leg-up or make them more efficient?

Horner: Well for me, I’m getting really jaded. I’m tired of hearing “I have an app for that.” I’m experiencing app overload. I’m kind of reversing thoughts, back to making it simple.

When it comes to solutions, if you can shift the time and complexity away from the end user, make it as simple as possible and most importantly, know and understand your customer. I can’t believe the complexity of some of these devices in connected health that they’re sending home with elderly patients. Really? You’ve got to be kidding me. Keep it simple and affordable.

Forget about trying to capture the whole supply chain. Find your place on the supply chain and just focus on it. Do the best you can there. Play nice with your competitors because the consumer is going to piece-meal their solution together. Apple has done this, but they’ve kept to their own ecosystem, so kind of do like Apple but in an open way. I’m a big open source advocate.

Campbell: Well that’s the topics that I had looked to cover. I appreciate you taking out the time. That personal story you told was so compelling. To me it’s stories like that, where you, the information technology leader for the organization that’s trying to improve that care, has the first hand perspective – a perspective that a majority of the time is lacking

Horner: That was my silver lining. That was my experience and I appreciate having gone through that. Because that was everything, from imaging to lab work up the wazoo, blood work every other day, the diet and dealing with the nutritionist. These providers, they want you to get better and they come up with their plan for you. These people come up with their plan for you, and it’s so much you’re overwhelmed. You were sick and recovering to begin with and then to be burdened with so much stuff to have to do. It’s almost like that’s the formula for you to fail anyway, even with those great intentions behind it. Somethings got to be there, that’s in my house, with me by my side—and it’s not my wife—that will keep me moving along but not somethings that’s going to overwhelm me. It’s got to be simple and easy. Part of my routine in the morning, is getting up, stepping on the scale. I do that already. It represents a great opportunity to capture that data because that’s got clinical implications. For others, there would be other things that they do, maybe someone gets on a StairMaster every day. Well do you hold onto it? Is it capturing your heartrate? Data. Capture it. Use it. Internet-of-Things, where we’ve got devices communicating to devices communicating to devices, intelligently. That’s going to do things. “Oh your heartrate is going up quite a bit, let’s lower the temperature in that room” I didn’t have to do anything. My smart watch already communicated to my smart thermostat and did that for me.

This interview has been edited and condensed.

Boston Google Glass Challenge

Google glass is a ground breaking technology that will revolutionize the way that physicians practice medicine. Last Wednesday April 23rd I was present for the Google Glass Challenge “Final Smack down” at Google in Cambridge, MA. I gained a new appreciation for Google Glass after viewing innovative demos of the new technology and hearing first account experiences of the technology from physicians.

The first speaker, Nayan Jain, Presidential Innovation Fellow told an amusing story that demonstrates how new the technology is and how it is not widespread yet. He spoke about being in a foreign country and how the government officials of that country thought he was a terrorist because he was wearing Google Glass. He said was hoping to show the officials how glass can be utilized for homeland security, but they were unimpressed. In other words, he said “he was lucky to get out with his life and his Google Glass intact”.

Rafael Grossman, a pioneer in medicine, the first physician to use Google Glass during surgery presented next. He started by presenting the shocking figures that 440,000 deaths occur every year from preventable medical errors and 40 times a week wrong side surgery occurs. Google Glass would rectify these hazards by notifying the physician when they were about to perform a medical error. Google Glass would also engineer new medical training techniques as med students would be able to see exactly what the professor was seeing from anywhere in the world instead of the dated method of students watching through a viewing window. Currently, we utilize EHR’s in the same way we did with paper records, clinicians manually put data into the EHR and take it out. Google Glass would allow information to be sent and received from the EHR automatically.

Chris Coburn, VP of Partners Healthcare talked about the upcoming disruption in the field of healthcare. Currently there are 6,000 hospitals but in the coming years there could be 60 or 60,000 hospitals as the delivery of care changes. He said that while healthcare venture is down as is the funding for lab science, a wider range of companies including hospitals are becoming involved in ventures. Hospitals are starting to back technology and having their own investments.

Dr. Karandeep Singh of Brigham and Women’s Hospital explained that, historically providers would query the patient for information about how the patient was doing. However, the invention of EHR’s has changed this interaction. The science of medicine has been preserved, but the art of medicine has been lost as providers now spend most of their time with the patient entering information into the EHR. Google Glass would allow a provider to stay at the patient’s bedside while reviewing information from the EHR and bring back the lost art of medicine. Amazingly, the technology is able to tell where the provider is in the hospital and automatically pull up the patient’s info based on which room the provider is in. Dr. Singh did an exciting demo where he used phrases like “show me the lab results”, “show me the trend for sodium”, “show me the medications”, and “show me the chest x-ray”. All of the commands swiftly provided the information that Dr. Singh had asked for.

Steven Horng from Beth Israel Deaconess Medical Center described how the hospital has employed Google Glass in their workflows. He compared the implementation of Google Glass to the implementation of iPads at BIDC. BIDC went live with iPads immediately after they released which allowed physicians to spend 30 minutes less at their workstations. BIDC went live with Google Glass three weeks after they received Glass and it is expected to decrease time at the workstation even more. Their V1 one of the deployment permitted information retrieval, which included lists of patients, vital signs triage notes, provider information with pictures, comments on what is happening with the patient, medication reconciliation, problem lists and allergies. The purpose of their V2 is to integrate Google Glass with the entire enterprise. This will use machine learning to use all the available data about the patient and determine how sick a person is based on mortality rates.

Google Glass is a very exciting technology that will transform the way doctors practice medicine and utilize their Electronic Health Records. It will allow providers to focus more of their time and attention on the patient which will allow the art of medicine to be reawakened.