Nearly two years after its passage, the Affordable Care Act (ACA) continues to provoke intense reactions. Proponents see it as the best chance for transforming a broken delivery system. States fear that provisions such as Medicaid expansion and the establishment of health insurance exchanges will break an already overextended system. The law's success or failure largely depends on the efforts of people like Virginia's David Mix.
As the program manager for Virginia's Health Information Technology/Medicaid Information Technology Architecture program, Mix is responsible for updating state IT systems that will support health exchanges and Medicaid expansion in 2014. Mix's background prepares him well for high-tech deep dives: He moved to the Commonwealth of Virginia two decades ago, after serving in the Navy as a nuclear submariner. I recently spoke with Mix about the dangers and opportunities ACA poses to state IT systems in this edited and condensed transcript.
Let's start by talking about the impact of ACA on Virginia's Medicaid program. The state is projecting that its Medicaid program will grow by 35 to 45 percent in 2014. What are the implications of that for Virginia's IT architecture?
The basic problem is: How do we handle that? We saw two ways to do that. We could either expand our current systems or replace them with something more modern. We found that if we expanded our current systems, then we would also have to expand our operational staff by roughly about 40 percent. That fixed cost works out to somewhere around $40 million a year.
Whenever state Medicaid directors gather, you hear horror stories about legacy systems. One Medicaid director even told me about working with a system that her father had helped install a generation earlier. Could you talk about the challenge of working with legacy systems?
The technology is 20-plus years old. Some of the systems have been rebuilt within the past 10 years but it's all archaic architecture. Our social services department, which runs the eligibility systems, is running on a Unisys platform using unsupported software and operating systems. We've been unsuccessfully trying to get off of that for many years.
If you're looking at it from the user's perspective, they have many systems they have to sign in for. They all operate differently. They have to know which system to use for the different categories of aid. It takes a lot of effort to get anything done, even simple stuff. These systems just don't talk to each other. They certainly don't line up with the vision of where CMS [the Centers for Medicare & Medicaid Services] wants to take Medicaid.
Did you decide to extend the current systems or build something new?
Extending current systems is really not a viable approach. We have got a lot of funding opportunities from the federal government right now. There is so much available that it gets kind of confusing. So what we did was we decided to use MITA [the Medicaid Information Technology Architecture] as a strategic direction and vision of what this should look like when we're done.
Describe that vision. What is MITA?
It's putting together a common physical architecture where somebody can have the capability for self-directed services. We have been working on what we call "No Wrong Door" for some time, but the scope of that effort was not from an enterprise perspective. Most of our current business processes are an assisted service model, where somebody goes into a facility and a state worker helps them [find and qualify for the appropriate type of aid]. If customers need assisted services, then they can either access a help center or they can go into a state facility to conduct their business. In some cases, they would have to go to a state facility, like for a driver's license. But for others they may not have to. What we are looking to do is provide a self-directed service model from an enterprise perspective. It's not something we offer today in the health and human services area.
You talked about the costs of continuing to use current systems to implement health-care reform. But building the type of common architecture and vision proposed by MITA is surely an expensive proposition as well.
That's true. But what's interesting is that if you just leave the programs silos alone, you start seeing some components of that service-oriented architecture being procured. All of a sudden, we've got multiple systems that are just specific for a program. If you look at it from an enterprise view, you buy one set of products that everybody can use. Your total cost will be lower than if everybody went out and just bought pieces of it.
What is the cost of trying to do this?
That's roughly around $17 million [to put] the infrastructure in place. Then you've got the systems work on top of it. Replacing eligibility systems is on the application side of that. A rough estimate of that cost in Virginia is $80 million to $100 million. Some states are a lot further along in their eligibility systems than our system. We are currently working in deploying a transfer system from the State of Michigan. It's a government-owned solution that will be the basis of our eligibility system.
What appealed to Virginia officials about the Michigan system?
The transfer system was selected as part of a competitive procurement for a new child care application. In looking at the other components in the transfer system, it fits with CMS's seven conditions and standards for funding. It's a Web-based environment, with a citizen portal where you can come in and apply for any of the social services programs. State workers have a worker portal where they login at one place. They can do their entire job within that environment and not have to go out to a dozen different systems. It also allows us to automate a lot of the fact checking job behind the scenes, both at the federal and state level.
Another nice thing about the technology is we can hook up legacy systems to it. When we took a look at whether or not we could hook up our current eligibility environment, we found that trying to coordinate across new technology would have cost probably half of what it would cost to replace the systems. With this system, once it's up and running, we can retire a dozen different older systems and get off those old platforms. They'll just become a memory and history. That's what we want to do, because it will help reduce our long-term operational costs while providing better service to taxpayers.
Has it been hard to get multiple agencies to buy into an enterprise framework that no single agency would control?
It is. We are collaborating with our transportation department. The Department of Motor Vehicles is a partner on this project. They are actually putting together the authentication service component for the enterprise. We are closely collaborating with the Virginia Information Technologies Agency, and talking to the secretaries of education as well as commerce and trade. I think once we get enough critical mass together, this will pick up its own steam and we won't have to press anybody to use it.
Even with federal support, it sounds like health IT, Medicaid IT in particular, is an area where state policymakers are going to have to do some investments at a time when investments are difficult.
That's quite true. But with all the federal funding that's on the table for the eligibility systems and everything else, it would cost the state more money in the future if we do not take advantage of the funding opportunities presented at this time. As for the budget problems, I view those as an opportunity because it's forcing us to innovate and think out of the box. We've got a lot of discussion across multiple program silos where we can collaborate and build something: a physical enterprise infrastructure that everybody can use and start saving all federal, state and local government programs time and money while providing better services to the taxpayers through the self-directed service model.
The Affordable Care Act gives states a choice between setting up their own health exchanges or allowing the federal government to set up exchanges for them. States that didn't like health reform to begin with were even more loath to let Uncle Sam make such important choices instead. However, a recent story in The Washington Post suggests that the dynamic is shifting in subtle ways. So many states have failed to take the measures necessary to set up exchanges that the federal government may not have the people and expertise to fill all the gaps. Which states are in the process of or have already established exchanges? The Commonwealth Fund offers this handy map.
The winners and losers of the great federal grant-stakes to establish exchanges are becoming clear. Last month, I shared stories on how California just picked up a $38 million exchange-planning grant. Meanwhile, Kansas is struggling to save a Medicaid eligibility modernization project, after recently rejecting $31.5 million in federal funding.
The Washington Post wonk-in-residence Ezra Klein predicts that Medicaid will escape deep budget cuts from the new Congressional Joint Select Committee on Deficit Reduction. Meanwhile, the Centers for Medicare & Medicaid Services (CMS) has rejected Illinois's efforts to tighten Medicaid eligibility requirements. The CMS-state negotiating show goes on! (Hat tip: Avik Roy.)
The Affordable Care Act is expected to reduce the number of uninsured patients by 70 percent. However, that could still leave about 20 million people uninsured. Wake Forest University professor Mark Hall and colleagues argue that safety-net programs could cover them for 25 percent less than the cost of Medicaid.
A quick look at health care innovations: Pennsylvania is considering paying Medicaid recipients to visit higher quality, lower cost providers. Georgia is pushing ahead with rural telemedicine. (Hat tip: Health Business Blog.)
Finally: New America Foundation's Shannon Brownlee and Joe Colucci take aim at "magic bullet thinking" in the health policy world. Their point? Policymakers and advocates must consider reforms (plural), rather than waiting for one magical reform.