In the second part of a two-part article about the challenges CIOs face in building clinically integrated networks, several health IT leaders share their perspectives on the critical role data analytics plays in the shift to value-based care and the need for collaborative leadership moving forward.
In part one of the story, published last week, George Conklin, CIO at the Irving, Texas-based Christus Health, a 60-hospital integrated healthcare delivery system and Mary Alice Annecharico, senior VP and CIO at Henry Ford Health System, a five-hospital health system based in Detroit as well as Tonya Edwards, M.D., physician executive at Impact Advisors, provided a look at building clinically integrated networks from the lens of a CIO and the challenges they face.
Healthcare Informatics Assistant Editor Heather Landi interviewed Conklin, Annecharico and Dr. Edwards following the Scottsdale Institute’s Spring CIO Summit in Arizona, in which 14 CIOs from leading healthcare organizations convened to discuss the most important health IT-related challenges facing CIOs. The Summit was hosted by the Scottsdale Institute, a Minn.-based not-for-profit membership organization of health systems advanced in IT, and sponsored by Impact Advisors, a Naperville, Ill.-based healthcare IT consultancy and moderated by Ralph Wakerly of Minneapolis-based consultancy C-Suite Resources. Insights from the discussions at the spring CIO Summit are outlined in the report, “Creating Clinically Integrated Networks: Challenges, Successes, Lessons Learned.”
Driven by the accelerating trend toward alternative payment models that reward quality of care rather than volume of services rendered, many of the organizations represented at the Scottsdale Institute CIO Summit have been preparing for value-based care with the development of clinically integrated networks for some time, while others are just getting started. Last year’s passage of the Medicare Access and CHIP Reauthorization Act (MACRA), which rapidly accelerates the transition to value-based payments, has especially spurred health systems to optimize and expand their clinically integrated networks, which presents CIOs with a number of IT challenges.
Conklin, Annecharico and Edwards discuss many of those challenges and lessons learned, and excerpts of the second part of those discussions are below. The interviews have been edited for length.
One of the key findings of the report was data analytics is the key to the kingdom, what does that mean?
Annecharico: We have struggled for so many years in the industry trying to cobble together inputs from all these different systems that we use and without regard for how do we sanctify that data, how do we master the data, so there is one true source of data—and that may be your financial data, your quality data, your clinical data, or could very well be your provider data—and then pulling it together and normalizing it, so that when we talk about a length of stay, or we talk about an event of care, we’re all talking about the same definition of that data. So, once organizations get to that point, we begin realizing that operations have to be bound by quickly turning over constant data. It has to be readily available and not staid, not two months old, in order to make good clinical decisions, good business and good strategic decisions, and in order to help us understand where our populations are and where we need to grow the business, or where we need to shrink the business. That’s absolutely vital to our organizations. It also help us with the measures that [the Centers for Medicare & Medicaid Services] and other regulatory bodies are looking for in terms of our quality outcomes and our cost performance. It will help us to keep the doors open, but it will also help us to conform across our systems, to a single standard, and an availability of data that now is really driving the business.
Mary Alice Annecharico
And, the CIO, as well as other executive leaders within an organization, is no longer going to be successful based on their personalities or the experience that they bring to an organization, they will be judged solely, like an organization’s health will be judged, by the availability and the agility that they can utilize data to help drive decisions and help with the business strategy of the organization.
Conklin: The consensus is that having more and better data is going to help us to ensure that we’re better able to deliver on our mission to provide high quality healthcare to all comers to our organization. But also will allow us to better evaluate markets and be sure that we put our community-based entities in exactly the right location. So, analytics help us make better decisions from a business perspective, and helps us to make better decisions from a clinical perspective. And that’s an obvious one, so when you appear at one of our free-standing ERs and we collect all your data and find out you have an allergy, and then you show up a clinic or acute care hospital, it ensures that we know about that information up front and are able to build that into the treatment plan that we create for you.
What is the sense of the progress that organizations are making in the area of data analytics?
Edwards: Progress around data analytics is all over the board. Some folks are just starting, and some healthcare organizations, particularly those that already have in place insurance arms, are much further along related to analytics. There are really significant challenges in healthcare around analytics for several reasons. First, we don’t share information well in healthcare, and trying to integrate information is difficult. The governance and data normalization practices are extremely challenging with healthcare. We have a lot of heterogeneous data with a lot of data that we may ultimately have to try to analyze, using natural language processing, because it’s not structured data. From the financial side, it’s fairly easy to analyze data. On the clinical side, it is not easy at all. It’s coming in many different shapes and forms. Providers may document actually the same thing in five or six different ways, so there’s standardization of processes that needs to happen in order to document in the same way so we can pull information in the same way. And then you may have challenges with laboratory data, for instance, where hemoglobin A1c that is performed by LabCorp looks different than A1c that’s performed by Quest or others, and being able to compare apples to apples. I’d say the challenges in analytics are around integrating non-standard data and data governance and normalization of that data so that then you can even begin to attempt to analyze the data.
Tonya Edwards, M.D.
The second challenge is that traditionally in healthcare we don’t have the types of resources that we need in order to be able to analyze that data well. For instance, it’s a new idea to have data scientists within healthcare organizations. We’ve very much been about analyzing retrospective data and reporting, essentially, and we’re just now beginning to move into being able to look forward into predictive and much less prescriptive analytics.
Another challenge identified in the report was demonstrating return on investment. Why have organizations found this challenging?
Edwards: The biggest challenge there is that CIOs and healthcare leadership teams have a tremendous number of demands, everybody wants capital dollars, operating dollars, and we’re at a time of shrinking margins for most healthcare organizations or systems. Those scarce resources are going to get allocated in areas where you can prove a return on investment. So many CIOs have, with the implementation of electronic health records (EHRs), for instance, where it was a situation where they anticipated significant ROI that may not have been realized, and there was not as much value out of the EHRs as was anticipated. It’s more difficult to prove, to bring value, and you really have to get in front of healthcare leadership and do some small projects that really do improve efficiency or save dollars or improve patient care in a way that, ultimately and indirectly, saves dollars or increases revenue in order to prove the value and be able to move forward into larger projects. It’s just a basic need at this point where we have to use our scarce resources in the best way possible.
Can you give some examples of how organizations are focusing on low-hanging fruit to demonstrate ROI?
Conklin: From a low hanging fruit perspective, how do we create and establish a long-term relationship with you as a person, something that’s “sticky,” which makes you want to come back and get service from us. So, part of that has been, historically, our service mentality and the models of care that we have built and developed that are very focused on the person, the needs of the individual, beyond just the medical care that’s given. What are you all about, what are your needs? We build a care plan around all those needs, and so create a relationship with you that is very “sticky.” That’s easy to do and inexpensive to do. We’re working now to actually build that out and we currently have a contract with a company that’s providing phone consultation services so patients can call them up and speak to a doctor and get some basic input, or a prescription. We’re in the process of building out that functionality and capability for ourselves. It’s relatively inexpensive to do that, and that’s low hanging fruit, and makes people want to come back for service again.
With this work of building clinically integrated networks, how is the leadership role for CIOs changing?
Annecharico: It will require more collaborative leadership. And it’s not all about the CIO. The CIO will be at the table to think strategically about the end point. But if we are really looking at clinically integrated networks, it’s a series of CIOs who need to be at the table and need to lead, inform and guide senior leadership in what does the data do and what does the data mean to help them be all in the same place.
For the six health system assembly that we are involved in here in the state of Michigan, we have an advantage and it was more coincidental than anything, as everyone will be in the Epic environments. The data sharing capabilities are already enhanced because we have the means of being able to do that in the Epic environment, but there will be the need for us to aggregate our claims and contract data in a way that will be really helpful. I think the CIO is a key part of it, but I also believe that all of the data-driven outcomes are not always within IT, and here they are not. I will tell you that we have an operations group that does data analytics as well as the population health analytic work; we provide the infrastructure and the end user tool that our leaders and clinicians use to interpret the data, but it is not run or managed by IT. I think more and more, we are realizing that there is a business strategy that is starting to help us shift as we become more and more data-driven healthcare organizations.
Conklin: I’m a little bit different from other CIOs. I come from a clinician perspective, as I’m a psychologist. The future is going to be about the CIOs, people sitting in my seat, who are going to have to understand and speak the language of the people running the organization as well as delivering the clinical care. They can’t be technicians, they can’t go to a meeting and talk in techno-babble, that’s a big turn off and people will not think of you as a partner. Usually when you bring up technical stuff, it’s brought up as an obstruction, such as why you can’t do something. What the CIOs need to be doing is not saying, “yes, but,” but rather, “yes, and.” The CIOs need to be talking to people about “This is what I can give you” as well as “I will give you want you want and even much more.”
And, absolutely yes, there is a need to be more collaborative. If CIOs approach a job primarily as a technician, we’ll get relegated into a support service very quickly, and we’ll lose any strategic standing within the organization. They have to be focused around the value that they can return, they have to be seen as an essential member of the leadership team and they have to contribute beyond what their particular area of expertise or interest is.
Edwards: Collaborative leadership has been a key for organizations that have been forming clinically integrated networks for some time. What we find is some of the historic leadership structures, hierarchical structures, don’t work as well because we have many different areas of expertise that are really needed. We find that it’s not just business operations that are leading these practices, we have to have IT leaders and clinical leaders, especially chief nursing officers, chief medical officers, CMIOs or CNIOs, who are informing some of these relationships as well as other operational leaders. There will be new relationships with chief analytics officers, chief quality officers and others. And they will need to work collaboratively because all these areas of expertise are important as we manage data across the continuum. We have to understand how are we going to use the data, what are the most important processes to apply the data to in order to make strategic changes that are going to improve efficiency, improve patient care and improve patient access. You can’t do that without these key leaders working together to understand what’s going to be the best use of resources. There is just a tremendous number of matrix relationships that become very important, as the senior leadership team works very closely together, across silos, rather than within silos.