Medicaid Learns from IT’s Mistakes

A case study on Infrastructure in Services

Executive Summary

In January 2005, the state of Maine launched its new, Web-based Maine Medicaid Claims System to process Rs 6,750 crore in annual Medicaid claims and payments. The new system was hailed as more secure , and able to clear claims faster, tracking costs better. But within days of turning on the new system, Craig Hitchings knew that something was seriously wrong.

 

There had been problems right from the start - an unusually high rate of rejected claims - but Hitchings, director of information technology for the state of Maine's Department of Human Services (DHS), had assumed they were caused by providers using the wrong codes on the new electronic claim forms. Hitchings' team - about 15 IT staffers and about 4 dozen employees from CNSI, the contractor hired to develop the system, - was working 12-hour days, writing software fixes and performing adjustments so fast that Hitchings knew that key project management guidelines were beginning to fall by the wayside. And nothing seemed to help.

 

The state's IT managers reasoned that a new end-to-end system would be easier and cheaper to maintain. The development of the new system was assigned to the IT staff in the DHS, which decided it wanted a system built on a rules-based engine so that as Medicaid rules changed, the changes could be programmed easily into the system. Some service providers offered states the opportunity to outsource claims processing systems. But the DHS staff believed building its own system would give it more flexibility.

For the next two years, CNSI and Maine's DHS IT shop worked long hours writing code. Errors kept cropping up. Hitchings and his staff made the decision to go live in January 2005. The switch to the new system would be made in a flash cutover in which the legacy system would be shut down for good and the new system would take over.

 

On January 21, Hitchings arrived at his office to find the claims system up and running. The initial reports from the contractor and his staff were that the system was humming along, quickly moving through Medicaid claims. But the following Monday morning, Hitchings sat down with CNSI contractors to go over the file statistics for the system's first three days. Something wasn't right. The system had sent about 50 percent of the claims - 24,000 in the first week alone - into a 'suspended' file, a dumping ground for claims that have an error that is not significant enough to reject the claim outright but that are not  accurate enough for payment.

Rebecca Wyke, head of Maine's financing department, appointed Thompson as CIO in late March, replacing Harry Lanphear. The goal was to get the new system to process claims at the same rate that the legacy system had, sending 20 percent into a suspended or rejection file. Thompson hired Jim Lopatosky, an Oracle database specialist in the state's Bureau of Information Services, as operations manager to act as a calming influence on the department's battered IT division. Lopatosky prioritized tasks. He acted as a liaison between teams working on different functions. He directed the programmers to fix those software bugs that would resolve  the largest number of suspended claims and postponed work on the portal, through which providers could check on the status of claims. That could wait. But the intricacies of the Medicaid program continued to thwart its progress.

Thompson needed a business owner who could clarify Medicaid business processes for the IT staff. Last October, Dr Laureen Biczak, the medical director for MaineCare, agreed to take on that responsibility. "This is what brought it all together," says Thompson. "It was something we should have done from the start: have someone who knew the business [of Medicaid] working full-time on the project."

 

The Person Behind It

image description
Jim Lopatosky
IT operations manager for the state of Maine
The IT staff was “running at 100 miles per hour trying to fix every software bug.
image description
Dick Thompson
CIO, state of Maine
By March, it was clear that we were missing any sort of basic management and were in a defensive mode.

Other Infrastructure Case Studies

image description
Prashun Dutta Senior EVP (IT & Quality), Reliance Infrastructure

Delivering IT at the Speed of Light

A case study on Power Consumption / Management in

It was a tall order. For Reliance Infrastructure, repairing cut cables quickly and creating power infrastructure within 30 days of a customer request wasn’t child’s play.But its IT head knew that only IT could deliver speed.

Other Services Case Studies

image description
Klaus Schelp Special Broadcasting Service (SBS)

SBS Looks to Cloud Services

A case study on Cloud Computing in Services

As the head of information technology at the national broadcaster, Schelp was met with increasing business demands for online collaboration without the on-premise infrastructure to support it. A journey to Cloud services looked increasingly compelling.