|
Videos
May 12, 2011 By David Raths
How old is the Medicaid Management Information System (MMIS) in Illinois? “Our joke is that the grandchild of someone who worked on creating our MMIS is now in long-term care,” said Ivan Handler, who recently left his position as CIO of the Illinois Department of Healthcare and Family Services to become CTO of the Governor’s Office of Health Information Technology, where he will work on the state’s health information exchange.
Traditionally MMIS has been a single mammoth system that handles a series of functions related to Medicaid claims processing and administrative information retrieval. But Illinois’ legacy mainframe system is buckling under the weight of ongoing federal health reforms — and it must be replaced soon, Handler explained. The longtime employees who have worked on the system are starting to retire, and it’s becoming increasingly difficult to find COBOL programmers to write “Band-Aids” to provide new functionality, such as a national provider identity requirement.
Also, deadlines for a host of federal mandates loom, like updates to the Health Information Portability & Accountability Act (HIPAA) and a new clinical and reimbursement coding system called ICD-10. The creation of state health insurance exchanges means eligibility systems must be updated. These changes will require greater interoperability between systems. The mainframe-based systems of the 1980s do not easily interface with each other or with data warehouses, which makes analysis and reporting cumbersome, if not impossible.
Illinois is by no means alone. At least 10 states responding to e.Republic’s 2010 Digital States Survey said replacing or upgrading their current MMIS is a top priority.
But Handler is convinced that the old business model of procuring a huge system that costs upward of $100 million and takes six years to implement won’t work considering the pace of health reform change.
“We don’t have six years to wait,” he said. Illinois has to be prepared to handle a half million new Medicaid enrollees by 2014. For that matter, the model in which states possess their own data centers also is outmoded, he added. “We don’t need to be running mainframes. We need to move to a cloud utility model. I want to look at acquiring business services hosted in the cloud instead of procuring and running our own systems.”
Handler would like to have the state health information exchange, hosted by a commercial service provider, house data repositories for different state programs. MMIS primarily involves claims processing, but he envisions a data warehouse supporting business services in a Lego-block kind of architecture. “We want to structure it so it can be built incrementally,” Handler noted, admitting that there aren’t vendors offering the type of services he’s looking for — yet. But he is pushing the half-dozen vendors that traditionally service the MMIS space and perhaps some newcomers to think outside the box.
Handler is currently working on an “MMIS and the cloud” presentation for an upcoming Medicaid IT conference. He would like more IT executives to consider a similar approach. “I wish more people would talk this up,” he said. “If CIOs rely on vendors for ideas, they won’t get the change they need.”
You may use or reference this story with attribution and a link to
http://www.govtech.com/pcio/New-Approach-to-Medicaid-Management-Information-Systems.html
Collaborative Justice: Transforming Criminal Justice Services Through Unified Collaboration
Cloud-Based Services Accelerate Public Sector Adoption of Video Collaboration
I think that many people are misguided by the tag ‘Legacy’. While the software may be old the operating systems and database engines that run on mainframes are not old. As a matter of fact DB2 runs circles around any server based system in speed and reliability. I have been told, since the advent of server technology’ that mainframes are a thing of the past, that was 20 or so years ago. This notion that you have to get rid of mainframes is non-sense. The idea of 'cloud computing', which is little more than a new way to say outsourcing, being the solution to this problem is equally non-sense. I believe the way to go is to properly utilize mainframes and use them as extremely fast and reliable file servers. The server environment is not conducive to processing large databases and I think you all know this. I have been in the business for over 30 years and I can count one 1 hand the number of times I have seen a mainframe 'go down' other than scheduled maintenance. Can any of you say the same about the server world? In short, the mainframes are already there, they are already being paid for, you are not going to be able to 'get rid of them' any time soon, so utilize them and take advantage of what they do good and shift what they do bad (user interface) to the server world. It will not only save money in the long run but will provide faster and more reliable computing. I encourage any of you to explain to me how I am wrong, please I love a good argument and I love even more ideas better than my own.
About thiws comment '“MITA is pushing states toward a more component-based system and away from mainframe architectures,” said Dan Sisco' Mr. Sisco, are you trying to assert thet component based or modular systems and the mainframe are mutually exclusive? Really? Then I assert you don't understand the concept.
My apologies, Should always proof read before hitting the submit button. About this comment '“MITA is pushing states toward a more component-based system and away from mainframe architectures,” said Dan Sisco' Mr. Sisco, are you trying to assert that component based or modular systems and the mainframe are mutually exclusive? Really? Then I assert you don't understand the concept.
Well...maybe not mutually exclusive but I think modular systems are much more difficult to develop within a mainframe ecosystem. And all the ancillary/3rd party applications are fairly hard to integrate. To me, Java and .Net apps, web services, etc. are the way to go. Lot's of "Sisco/Cisco/Sisko's" commenting here. :')
I think the biggest problem is that systems like Oracle, Tera Data, SQL Server et al, cannot handle very large databases and they cost more to boot. My model is to use the mainframe (if you alredy have one) as a very fast file server with DB2 as the database engine. The only programming that need be done on the mainframe (in this model) are the batch updates and other maitenance. The Business rules and GUI applications would still be (and should be) in the server environment. it is a simple solution to many complex problems and I don't see why more IT people do not recognize this. The only reason I can see is that the people making the decisions do not have a real grasp of the problem and that the people they do listen to are either the vendor sales people or their own IT people who are just enamored with new rather than right. PS - The e-mail I am giving here is to a guy who has done extensive work on the geneoloy of the Cisco/Sisco family tree phsisco@gmail.com
From the ICD-10 deadline to the need for medical scribes, the hurdles facing EHR implementation are numerous. http://www.healthcaretownhall.com/?p=3858
Just think of mainframes as servers on steroids. All these guys who learned the business in the past 15-20 years think this is new stuff. The fundamentals are the same. Keep it simple. Today's architects love techno-speak, elegant solutions with the latest technology and moving parts. Unfortunately, the moving parts usually mean up and down. Touch one part of these complex architectures and you'll understand how much fun it is to watch the dominoes fall. You could learn something from us "legacy" guys.