By Cindy Atoji
July 8, 2008 | Bringing Medicaid into the digital age isn’t easy agrees CMS’ Rick Friedman, who compares it to “turning the Queen Mary around.” As the federal government teams with the states to incorporate health IT into Medicaid programs, “the biggest challenge is not the silicon but the carbon; that is to say, it’s the people, not the technology,” says Friedman, who is director of the state systems for the Center for Medicare and Medicaid Services (CMS) with the U.S. Department of Health and Human Services.
Currently, the many state Medicaid systems constitute a costly tangle of diverse manual and computer operations that lack standardization and are poorly integrated, argues Friedman. MITA—the Medicaid Information Technology Architecture—is the blueprint for untangling this knot and moving from claims-based processes to “a service-orientated architecture (SOA),” he says.
Medicaid is the single largest budget item for most state governments, and “the opportunity to cut down on inefficiencies, improve services, and reduce fraud—all of this is out there,” maintains Friedman, who spoke with Digital HealthCare & Productivity about how states are using health IT, such as electronic medical records, ePrescribing, and other health IT to transform Medicaid. Jessica Kahn, a CMS project officer in charge of CMS’ $150 million Medicaid Transformation Grants, joined Friedman.
DHP: Why is MITA so necessary?
Friedman: For the past 30 years, Medicaid agencies and states had their own systems, usually handled by different contractors. These systems are homegrown from the bottom up because the policies in each state for Medicaid -- ranging from eligibility, benefits, reimbursement rates, and waivers -- vary from state to state. The IT systems reflect that diversity.
The difficulty is that these systems cannot share data easily across organizational silos, within states, or from state to state. So MITA is a transformational change. It is CMS’ effort to try to transform the transaction-based, claims-focused, point-to-point systems into a service-orientated architecture based that is based on business processes that are common to all states, and affords sharing across organizational boundaries.
DHP: How is the transition from the old state Medicaid Management Information Systems (MMIS) to MITA going?
Friedman: We are probably halfway through, in terms of the design of MITA and only 1/5 way through in terms of adoption of MITA by states. So we are busily working away, creating this thing called MITA, and it has a whole bunch of different parts to it. As state systems come up for renewal and change, they are taking bits and pieces. In some cases, they’re replacing in total; in other cases they’re taking a piece of it and working with it. Depending on how you slice it, we’re either halfway there, or 20 percent from the beginning in terms of implementation. It will take the next 8-10 years before the states have separated themselves from legacy systems and moved toward SOA.
DHP: What are the different MITA components?
Friedman: MITA consists of a framework, planning steps, and different processes that states are supposed to use. From an architectural perspective, MITA consists of three architectures: the business architecture; the information architecture, which is the data; and the technical architecture, which is the infrastructure to make it work. The business architecture is basically the idea that service-orientated architecture focuses on the business of a particular activity first, not the technology.
MITA’s business architecture has broken down discrete pieces of what all states have to do when they administer the Medicaid program, as diverse as they are. There are basically 78 different business processes—such as provider enrollment, etc. Business architecture sets the stage for rest of it. The technical architecture is primarily SOA orientated that you could find in other industries like banking or transportation. I think of it as if it were a Tinker Toy set. You can build different things with the Tinker Toy set because you’ve broken in down into such discrete pieces. In fact, two different states can build two different systems to meet their unique needs, but both based on the same standards. So that’s basically the idea behind SOA in the Medicaid context.
DHP: How does MITA fit into the landscape of HIT and HIE?
Friedman: States are wonderful laboratories in terms of being on the cutting-edge of real world health care for some of America’s most vulnerable citizens. And as a result, they are not hamstrung by a one-size-fits-all solution. States are doing a whole bunch of creative things in eHealth world and MITA is an important part of that, but the leadership is exhibited by the states.
Kahn: I’ll talk about the Medicaid Transformation Grant, a large percentage of which is focused on HIT and E [processes]. The Medicaid Transformation Grant are funds awarded to the Medicaid agencies to implement significant reform in how Medicaid provides its care, mostly focused on efficiency, cost savings, and care coordination. Even though it was put out as a broad solicitation most of the proposals that came in involved HIT and E. And it’s closely coordinated with MMIS because the initial input into what they are trying to exchange among the Medicaid providers is the claims history for the Medicaid beneficiary, and then, on top of that, they are adding clinical data, laboratory, imaging data and so forth.
There are about 25 states that are looking at this [and] they are all approaching it differently. Some are piloting electronic health records; some are focusing clinical decision support, others are looking to partner with private payers in their state to implement health information exchange while others are looking at this as Medicaid really driving the boat and bringing on others as they go along.
Friedman: The Medicaid program can serve as an architectural backbone to provide not only as a source of sustainable revenue but more importantly, an architectural framework for working with RHIOS and other groups.
DHP: How can MITA’s return on investment be demonstrated?
Friedman: There’s anecdotal information about the inefficiencies of the current system, relative to duplication, potential fraud and abuse. In the existing systems, one only gets to see a part of puzzle, and you’re sort of wearing blinders. To the extent that MITA becomes fully operational—and it isn’t anywhere yet—there will be a much broader view of data being pulled together. I think frankly, the jury is still out.
Kahn: For the transformation grants, the areas they’re looking for cost savings in health are reduced duplication of testing, improvement in preventive care, ePrescribing, reducing likelihood of adverse events, and other areas. These sorts of activities are going to take time to show a return on investment.
Friedman: Over time with the MMIS, the number of vendors or contractors who could build MMIS systems began to shrink until there were just a small handful of vendors from whom Medicaid agencies could choose as contractor to build the system. Because MITA breaks down business services into smaller discrete and common business services, it enables niche players to be able to play in the market space where they have not been able to get a beachhead before because of these huge monolithic systems that were all highly tied together and very difficult to change over time. So we’re envisioning that there is going to be a significant benefit relative to just increasing competition.
DHP: What obstacles still need to be overcome with MITA?
Friedman: I think funding is an issue, not only financing, which everyone appreciates is always a challenge, but financial policies that facilitate rather than hinder adoption of MITA. And the single biggest challenge is the leadership—people with vision and the fact that this is going to take a couple years, and going to last longer than next administration, governor, or division director. Those all things not unique to MITA but any enterprise that has some vision.
DHP: Do you envision that Open Source will have a large role to play in the direction of MITA?
Friedman: Yes, absolutely. Historically, there is a requirement that whatever is paid for in federal dollars is supposed to be out in the public domain. But the reality of transferring MMIS systems from one state to another, in spite of the requirement, has been very, very difficult, for a variety of reasons, not the least of which is the technology. And by the time you get done customizing it for the second state, you’ve run up a huge tab. With MITA, not only is it broken down into these discrete pieces of the Tinker Toy set that are replicable, but the concept of Open Source is absolutely critical to all of this. We want to do as much as we can to push in that direction.
Kahn: With the Medicaid Transformation Grant, they are either developing something themselves that they’re going to make Open Source or they’re using, for example, VISTA. They’re sharing programming issues and looking at creating a use case for a core administrative data set that would help enable the exchange across states.
DHP: What comes first, Medicaid reform or Medicaid Health IT? Is this a case of the chicken or the egg?
Friedman: There’s only one chicken, and it’s Medicaid reform; the business of Medicaid comes before the IT of Medicaid, so Medicaid reform is the driver. IT follows at a respectful distance.