David Williams: This is David E. Williams, co-founder of MedPharma Partners and author of the Health Business Blog. I’m speaking today with George Conklin. He is Senior Vice President and Chief Information Officer of Christus Health. George, thanks for you time today.
George Conklin: You’re very welcome David.
Williams: What is the size and scope of Christus Health?
Conklin: Christus Health is a 44-hospital Catholic health care system based in Dallas, Texas. We have hospitals in Texas, Louisiana, Arkansas, Oklahoma and New Mexico and seven facilities in northern Mexico.
Williams: What are your information technology capabilities?
Conklin: We have a large data center that we opened in November of 2008 in San Antonio, Texas and a backup for disaster recovery about 25 miles away. We operate our main and mission critical systems out of that data center, but have a growing number of applications that we operate in the cloud, Humedica being one of those. We have a smaller but growing data center in Monterey, Mexico for operation of our Mexican hospitals and as we extend further into Central and South America for those organizations as well.
Taking it up a layer, we connect over 380 different locations where Christus does business. They run the gamut from our 44 hospitals to doctors offices, clinics, home care agencies and a wide variety of other kinds of health care services that we operate. Also –as a reflection of our three-part business strategy—we operate a large array of non-acute services and retail services. These retail services are found in our hospitals right now but are beginning to move into shopping malls and other locations. We now also have a web presence; you can go to www.christushealth.org and branch off to commercial products that you could buy. You could buy vitamins and lotions and things like that on our website as well.
The three-part strategy includes: (1) drastically revamped acute care through a network of hospitals focused on delivering low cost high quality service, (2) non acute and retail and (3) international.
Our IT strategy is to aggregate information across all of those different entities. We want to be able to bring just the right information back to the point of need. So a physician treating you would be able to get just the right information to help me in my treatment of you. If I were an administrator looking at the class of all the David Williams’ or class of all people with congestive heart failure, I would be able to look at the best ways of treating people. The objective is to move us toward our goal of being a low cost, high quality provider.
Williams: That sounds like a very well thought out strategy. How close is it to being realized? Is it in place today or is that more of a long term strategic vision?
Conklin: That’s a longer-term strategic vision. Today, through our portals, a clinician could gain access to George Conklin’s information from anywhere, but would have multiple log in’s and multiple systems to look at. We are actively seeking a health information exchange engine to sit on top of all our systems.
We want it to do three things for us.
One is to present the information in a uniform fashion so you won’t have a Meditech system in one location that you’d have to learn how to deal with and a server system in another location that you’d have to deal with. Instead, you would just have one way that you interact with the systems. So the first thing would be the presentation piece.
The second piece is the decision support piece. There really are no HIE products today that provide that decision support component. We’re working with vendors now to glean just the right information for the episode. If you think about it, as health information exchanges begin to grow around the country, one of the big concerns among clinicians is being buried in information about David Williams and having to paw their way through all this extraneous information. They need to get to specific pieces of information. It’s important to have all that other information available, but there is also an awful lot of it that’s not going to be relevant to a particular care event that somehow or another we need to figure out how to pull apart. So decision support in part would help to parse the information so just the right information is being brought to the clinical event. But the second part of it is to really help a clinician make better clinical decisions based upon science, based upon the costs of different kinds of treatment, based upon the organizational protocols that have been set up for particular organizations like Christus Health. So that’s the second component of that HIE architecture.
The third component of it is data aggregation for large analytical studies that let you determine the best ways of treating people with diabetes, COPD and so forth. As a former clinician and psychologist myself, when I used to see patients in the emergency rooms, one of the questions I would ask is, “David, why are we here today versus yesterday?” One of the things that was driven home to me very early on in working in the community mental health movement is that there is a difference –and it’s not always obvious– between people who are sick and people who are ill. Most of our treatment decisions are based upon bodies of data aggregated in hospitals. This is not necessarily the best body of data for me to keep you well in the community. So part of our plan is to be able to integrate data across a very large delivery spectrum to be able to look at how we keep diabetics productive and happy in the community longer, not base our treatment protocols on patients who are in our hospitals.
Williams: Speaking of analytics, let’s discuss your work with Humedica. I’ve interviewed the CEO and some of the top staff there and they cite you as a key partner in the development and launch of what they’re doing. Can you tell me a little bit about you work with Humedica? What are the objectives? What benefits are you seeing?
Conklin: One of the principals in Humedica is a woman named A.G. Breitenstein. We have worked with A.G. in a number of different positions that she’s held over the years, every one of them having to do with managing large sets of information and deriving usable and actionable clinical or operational information from them. So when A.G. moved to Health Insight, the predecessor of Humedica, we worked with her on new products in development. We have been working with them over the last couple of years on the creation of the database and the analytical framework.
One of the reasons they wanted to work with us is that we did something back in the 2004 to 2006 time frame that many organizations probably wished they had done and are now going to have to move towards themselves. We undertook a massive information and clinical protocol standardization effort across Christus. We had remarkable acceptance from management, clinical leadership and clinicians across the organization around the concept of moving toward significant standardization. We undertook standardization of lots of clinical and business processes, ending up with literally tens of thousands of items of both clinical and operational information standardized across the organization. When we came to Humedica our ability to be able to pull essentially normalized data out of our systems and provide it to them so that we could create very large data bases gave them an awful lot of information that they could very easily work with as they were developing and testing their methodologies. We were peculiarly positioned because of those decisions that we made back in 2004 to 2006 and we continue to move forward with that model today.
We were able to establish a very rich clinical and operational database that let us easily marry a lot of information about patients with cardiovascular disease or diabetes, etc. across the millions of visits and hospitalizations that we see on an annual basis. That gave Humedica a very large database to work with without having to go through a lot of normalization effort on the front end. It allowed them then to focus on producing analytics that were very useful to us.
As you are aware from your prior discussions with Humedica, there are effectively two products that they’ve got. One of those is for retrospective analysis of information and the other is for near real-time or prospective analysis. The retrospective analysis has been utilized by us over the last several months in pilots to provide clinical leadership with detailed information on an aggregate basis as well as at a physician level about the performance of physicians in different clinical areas. It allows us to go back in those regions where the system is being piloted to look at the low cost, high quality providers, learn what those providers are doing and then talk to the other providers in that area about what we’re seeing from those other low cost, high quality providers.
That is beginning to slowly move performance from timeworn ways that people have been doing things toward newer, better practices. It confirms what we’ve believed for a long time, which is that if we provide physicians with data about how they’re performing then their behavior will change in positive directions.March 12, 2010