VA is a much more complicated rollout since there are so many different interactions and configurations of VistA. In addition,…
Sean Carroll is CEO of Arcadia Healthcare Solutions of Burlington, MA.
Tell me about yourself and the company.
Arcadia is an EHR data harmonization and analytics company. We focus on building high quality, highly usable data assets for risk-sharing entities such as health plans, IDNs, and IPAs. The scope of the business has us covering 20 million patients, 40,000 providers, and 4,000 practices, both owned and affiliated.
As for myself, I am a lifer in health IT. I’ve been at it for almost 30 years across six companies. All of those companies have had some principal focus on data and some form of disruptive technology or business model component. I’ve been here at Arcadia for two and a half years.
Your solutions connect to the back end of EHRs. Interoperability seems to have settled on two sides of the equation, one being real-time integration that requires vendor participation and the other being to extract information in some other way as needed. Do you see that perhaps the market forgets that external applications can sometimes access EHR databases directly?
Yes. That’s been our focus, certainly for the last decade — working quite deliberately on the back end of the top 30-plus electronic health record systems in the market. I think right now the standard is less about two-way operability, especially between EHRs. That’s very rare if not non-existent. But more so the kind of deep integration that is needed to execute against the kinds of measures that are emerging in the marketplace largely driven by value-based care.
Do you need the EHR vendor’s help to understand their data catalog and metadata or can you discover that on your own?
We don’t need their help, necessarily. We certainly need a customer who has invested in electronic health records to work with us to make all parties helpful to the process, because in the end, it’s the patient we’re trying to help, and it’s the customer who has made that investment who needs to drive how to get at that data to provide quality care and lower cost.
You connect to 30-plus data sources. How much information outside the EHR is needed to give you a complete picture of a patient or of quality?
Right now I would say it’s very helpful fringe-level data. Most of the market is still reconciling to the notion that deep clinical data from electronic health records is paramount to creating a high quality, highly usable data asset. We do have clients who are already well into that path, of course, and have asked us to pull in data from practice management systems or other systems that have bits and pieces of information that might not exist elsewhere.
What insights are customers discovering that they wouldn’t have been able to figure out just by looking at the EHR?
A simple example would be if you are looking at claims data — which is principally how people begin to think about analytics around healthcare data that’s been the standard for so long — you would be able to see from a claims component that someone had a cancer screening test done. But without the integrated EHR data in that analysis, you wouldn’t know necessarily whether they have cancer. If you think about where healthcare is trying to move to in terms of closing gaps in care and being efficient, the combination of those two things is what’s really needed to be more timely and efficient in how you handle the patients. That’s a very basic but I think a very important and high-profile example.
Providers often don’t know what questions to ask until they see a report that, by definition as a canned report, reflects the collective best practices of the vendor’s other customers. Are your off-the-shelf reports a surprise to providers who wouldn’t have thought about looking at specific information on their own?
Absolutely. Some of that is driven by the breadth of the information that results from that combined data set. But oftentimes with electronic health record data in the mix, you’re seeing things much more real time than you would from claims-based analysis only. They’re in a position to react to situation much more quickly through deeper and broader information that is much more timely, as most of our data refreshes every 24 hours.
EHRs focus on transaction management and data completion. They don’t do a lot on the front end with patient engagement and then on the back end some of them don’t have robust analytics. Do you see the post-EHR era being three legs of a stool with the EHR vendor providing just one?
Absolutely. The future would suggest that it’s the next generation of systems that have the capability to harmonize data from a variety of systems and draw insights from that aggregated data set. That was the original thesis for the electronic health record. Given how adoption has been less and it has taken the time that it has and the business model of value-based care and global payment is now in the driver’s seat in the marketplace, I see the electronic health record systems as a source of information among many. Albeit a very very important one and with a great deal of the necessary information, but still just a source.
EHRs were supposed to be different from EMRs because they would collect and present health information from many systems in many encounter locations outside a given provider, such as dental offices, drugstores, and long-term care facilities. That EHR concept was sidetracked when ONC decided to certify the same old EMR products and call them EHRs. Would you agree that no provider has deployed what might truly be called an EHR under that original definition?
There are unique deployments of electronic health records with unique organizations that have gotten close to the original promise of what they were intended for, but the vast majority of the market has not realized the original dream. Based on the slow march towards value-based care, we’re going to see a reset where next-generation technology is going to drop on that substantial footprint of EHRs that exist, but it won’t be the single answer. It will have to be compiled with clinical, business, and claims data from other systems to affect and support the change that’s required in the healthcare model.
Is it common now to incorporate claims data?
It’s more common. Certainly the payer marketplace is recognizing that their data coupled with clinical data is a great asset in the marketplace. About half of our clients are payers and some of the more advanced ones — like a large Blue Cross organization in New England that we work with — use aggregated claims and electronic health record data to support the administration of a very creative pay-per-performance program. That’s been very successful in bringing together providers and in the plan on the premise that if we share information carefully and appropriately, we can in fact provide incentives, control costs, and affect quality in the way that we want.
There are certainly real things happening out there with data when it comes together with the provider side of market and the payer side of the market. It works the other way, too. We have direct clients who are large provider organizations or large ACOs who are doing the same thing for similar reasons. But the concept is very much the same – the datasets together provide the lens into what’s happening across principally their ambulatory networks and they can see and manage at the population level.
Are providers are getting into the payer side of the business?
Sure. We talk to provider organizations all the time who are contemplating moving toward building a plan.
We see this in both directions, but the trend we’re seeing more is a much stronger willingness to come to the table, provided that the technology exists and there is the presence of some form of trusted third party — which is a role that we typically play — to help aggregate and arbitrage the right data to the right people in a very trusted and appropriate way. We’re seeing that trend more than providers standing up plans or plans somehow getting closer to providers.
What factors should a provider consider when choosing an analytics vendor?
It’s a very needed competency. It truly is all about the data when it comes to being effective in a value-based model. I would make sure that a supplier can connect you up with clients who’ve really put the technology to use and have seen tangible outcomes. Many organizations in the market are still early stage in the development of their technology. Secondly is the question of the source. The source in our mind is electronic health record data.
It’s very customary for us to engage in a dialog with even a medium-sized IDN who might have 50 different EHRs across their network. When you think about extracting the right data from 50 different systems just at the EHR level and getting that harmonized appropriately, it’s very heavy lifting. I would make sure that who you’re talking to can demonstrate that capability in a real way and with references.
The last piece goes back to the provider themselves. Do they have a clear strategy? Because what we’ve found is that many organizations know that they need to move in this direction and they know that data and technology in particular is important or perhaps even a backbone, but they haven’t fleshed out their full plan yet. Therefore, they’re not quite ready for the technology. That’s one of the reasons we acquired the Sage business — to help those organizations who are just a little more early stage to move closer to value-based or risk-sharing before making the investment in a solid data asset on which to drive the strategy.
How did the Sage Technologies acquisition change what you offer?
It added a deep tenure in managed care through this Midwest-based business that provides end-to-end services to provider networks that are engaged in risk-based contracts with managed care payers and ACOs. They provide everything from claims processing, network administration, utilization management including case management, customer service, data management, reporting, and critical care management. Really a full suite of supporting services that are required for an IPA or some form of other provider network to execute when they’re engaged in risk and to be good at it.
A large part of the market is still in that state, thinking about more aggressive moves and deeper risk arrangements where technology starts to become more critical. We wanted to have an ability to serve those clients now and also to make sure that we had the resident services to offer some of our technology clients in support of their activities. It has helped us with a little bit more of an end-to-end capability serving a larger portion of the market, which is very much in transition with a variety of different maturity levels amongst the organizations as it relates to risk-based contracting.
How would you like the company to change over the next five years?
We’re very dedicated to the notion that clinical data in particular — for the next five years and perhaps beyond — aggregated from electronic health record, is fundamental to an effective data strategy. A data strategy is fundamental to being successful in value-based care. We’re focused on that.
We certainly understand the necessity to deliver on the full outcome, but our focus will remain on solving this important and fundamental challenge that organizations have, which is, "I’ve made huge investments in my electronic health record strategy. I need the information out of all of them. I need it timely. I need to be able to then process it right it away in much broader ways, including looking at the full population that I serve. That’s the only way that I will be effective in executing in any sort of risk model."
Our focus will stay there. We hope to be the recognized leader in that particular competency. We’ve been at it for 10 years. We have quite a bit of intellectual property in and around that process. Beyond that, our mission is to help patients and help the system evolve in a high quality way and to deliver to providers a useful tool that will be efficient in the way they provide medicine as these models evolve.
Do you have any final thoughts?
We’re very enthused that the market is signaling clearly that value and value-based models are the landing spot. We see that through multiple things happening with CMS, including recent announcements about supporting value-based characteristics and Medicare Advantage. That’s just another signal. We’re very curious about that. We think that that is where healthcare should be. We think we can play a significant role in assisting in that journey.
Clinical data from EHRs is a difference-maker. We’ve seen it over and over again with our 40 clients. The speed, the depth, and the comprehensiveness of that data, coupled with payer data and other sources, is critical. We believe plans and providers can and will — and in fact, must — come together to share the kind of information that will make all this possible. We’re seeing that happen more and more in the marketplace. We’re looking forward to being a part of this tremendously positive momentum that’s occurring.