Old code proves key to modern IT at Midland
2008 InfoWorld 100 finalist: Hospital adopts and adapts 30-year-old code to establish award-winning automated records system
Follow @infoworldIn times of financial crisis, it pays -- or, more fittingly, saves -- to explore unconventional ways to pull off major technology projects. For Midland Memorial Hospital, this came in the form of 1970s-era code unearthed via the Freedom of Information Act.
Major executive shakeup and financial losses in the millions left Midland Director of IS David Whiles at a loss as to how he and his team could breathe new life into the Texas-based hospital's aging IT infrastructure. The comprehensive EHR (electronic health-care records) system Midland needed to modernize its operations typically cost around $22 million. Thanks to 30-year-old code, creative partnerships, and a deep commitment to end-user training, Whiles and company found a way to do it for only $7 million.
[ See the full list of this year's InfoWorld 100 winners, as well as the other Top 10 finalists ]
The journey began six years ago as most IT projects do: with a year-and-a-half search for the right vendor. A dozen vendors showed up on Midland's radar, says Whiles, but "we didn't see anything we felt we could afford."
Whiles knew there had to be a cheaper route to EHR, which the hospital needed to convert paper-based processes into electronic ones for increased efficiency, security, and compliance with new regulations.
That's when Whiles looked outside of the traditional EHR vendor box and found VistA, EHR technology developed in the mid-1970s and used mostly by the U.S. Department of Veteran Affairs (VA). Based on open source code, VistA was made available to Midland free via the Freedom of Information Act. The price tag for hardware, software, and services to support VistA was irresistible: about a third of the cost of other EHR systems.
But many challenges awaited Midland's adoption of VistA. For starters, the code needed to be tweaked to match Midland's best practices -- yet Midland didn't have any programmers on staff. Also, interfaces had to be built for third-party systems so that VistA could handle operations outside the typical VA facility, such as ob-gyn (babies aren't often born at VA hospitals), acute care, and surgery, as well as sophisticated registration and billing that works with various insurance agencies.
To do this, Midland needed an integrator to help port VistA to a private hospital. Hardly anyone offered such a service, says Whiles, because "this was never done before." Whiles eventually found Medsphere, an unproven startup with no customers. But the possibility of a partnership with Medsphere exposed Midland to significant risks. What if Medsphere proved incapable of handling the multimillion-dollar, multiyear project? What if Medsphere suddenly went out of business?
To assess these risks, Midland turned to Deloitte Consulting, which provided a plan to mitigate them: gauge Medsphere consultants' willingness to sign on as Midland employees should Medsphere go belly up.
Whiles reached out to Medsphere under those conditions in December 2004, and the rollout began. Midland clinical analysts and Medsphere technical pros collaborated deeply for the better part of two years to develop the necessary interfaces for Midland's VistA-based EHR.









